A computer-implemented method for ordinal prediction is provided. The method includes encoding time series data with a temporal encoder to obtain latent space representations. The method includes optimizing the temporal encoder using semi-supervised learning to distinguish different classes in the labeled space using labeled data, and augment the latent space representations using unlabeled training data, to obtain semi-supervised representations. The method further includes discarding a linear layer after the temporal encoder and fixing the temporal encoder. The method also includes training k-1 binary classifiers on top of the semi-supervised representations to obtain k-1 binary predictions. The method additionally includes identifying and correcting inconsistent ones of the k-1 binary predictions by matching the inconsistent ones to consistent ones of the k-1 binary predictions. The method further includes aggregating the k-1 binary predictions to obtain an ordinal prediction.
G06N 3/0895 - Weakly supervised learning, e.g. semi-supervised or self-supervised learning
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
Transmission methods and systems include calibrating a mismatch between uplink and downlink digital elements of hybrid beamforming transceivers. calibrating a mismatch between uplink and downlink analog elements of the plurality of hybrid beamforming transceivers. Respective downlink channels are estimated between a user equipment and the hybrid beamforming transceivers using respective mismatch calibrations for the digital and analog elements of each of the hybrid beamforming transceivers. Data is transmitted to the user equipment from the hybrid beamforming transceivers using a distributed beamforming pattern based on the estimated downlink channels.
H04B 7/06 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
H04B 7/08 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the receiving station
3.
ANOMALY DETECTION USING METRIC TIME SERIES AND EVENT SEQUENCES FOR MEDICAL DECISION MAKING
Methods and systems for anomaly detection include encoding a multivariate time series and a multi-type event sequence using respective transformers and an aggregation network to generate a feature vector. Anomaly detection is performed using the feature vector to identify an anomaly within a system. A corrective action is performed responsive to the anomaly to correct or mitigate an effect of the anomaly. The detected anomaly can be used in a healthcare context to support decision making by medical professionals with respect to the treatment of a patient. The encoding may include machine learning models to implement the transformers and the aggregation network using deep learning.
An AI-driven cable mapping system that employs distributed fiber optic sensing (DFOS) fiber sensing and machine learning that provides autonomous determination of fiber optic cable location and mapping of same. Designed Al algorithms operating within our inventive systems and methods provide an easy solution for cable mapping in a GIS system; automatically maps using landmarks and manhole locations; and employs a supervised learning algorithm. A vehicle-assist operation is employed wherein a vehicle carries a Global Positioning System (GPS) device and drives along a roadway thereby following the fiber optic cable route; data paring that provides further significant locational information wherein time synchronizes between the DFOS system and vehicle GPS device from which we automatically pair the data of fiber length from traffic trajectories and GPS coordinates by time series.
Systems and methods for out-of-distribution detection of nodes in a graph includes collecting evidence to quantify predictive uncertainty of diverse labels of nodes in a graph of nodes and edges using positive evidence from labels of training nodes of a multi-label evidential graph neural network. Multi-label opinions are generated including belief and disbelief for the diverse labels. The opinions are combined into a joint belief by employing a comultiplication operation of binomial opinions. The joint belief is classified to detect out-of-distribution nodes of the graph. A corrective action is performed responsive to a detection of an out-of-distribution node. The systems and methods can employ evidential deep learning.
G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
G16H 50/70 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
A data-driven street flood warning system that employs distributed fiber optic sensing (DFOS)/distributed acoustic sensing (DAS) and machine learning (ML) technologies and techniques to provide a prediction of street flood status along a telecommunications fiber optic cable route using the DFOS/DAS data and ML models. Operationally, a DFOS/DAS interrogator collects and transmits vibrational data resulting from rain events while an online web server provides a user interface for end-users. Two machine learning models are built respectively for rain intensity prediction and flood level prediction. The machine learning models serve as predictive models for rain intensity and flood levels based on data provided to them, which includes rain intensity, rain duration, and historical data on flood levels.
Systems and methods for manhole localization along deployed fiber optic cables that employs cross-correlation methodologies and ambient road traffic operating proximate to the manholes including fiber optic telecommunications cables to detect the manhole locations using distributed fiber optic sensing (DFOS). Advantageously the manhole locations are determined without employing labor intensive field surveys.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
8.
DYNAMIC LINE RATING (DLR) OF OVERHEAD TRANSMISSION LINES
Systems, methods, and structures providing dynamic line rating (DLR) for overhead transmission lines based on distributed fiber optic sensing (DFOS)/distributed temperature sensing (DTS) to determine temperature of the electrical conductors. Environmental conditions such as wind speed, wind direction, and solar radiation data, are collected from environmental sensors and an acoustic modem that convert the digital data collected from the environmental sensors into generated vibration patterns that are subsequently used to vibrationally excite a DFOS optical sensor fiber. The DFOS system monitors the optical sensor fiber and detects, measures, and decodes the vibrational excitations. An Artificial Neural Network (ANN) determines a heat transfer correlation between the temperature of the optical sensor fiber and electrical conductor(s) (core temperature).
A device may receive, from a fiber sensor device, sensing data associated with a fiber optic cable, the sensing data being produced by an activity that poses a threat of damage to the fiber optic cable, and the sensing data identifying: amplitudes of vibration signals, frequencies of the vibration signals, patterns of the vibration signals, times associated with the vibration signals, and locations along the fiber optic cable associated with the vibration signals. The device may process, with a machine learning model, the sensing data to determine a threat level of the activity to the fiber optic cable, the machine learning model having been trained based on historical information regarding detected vibrations, historical information regarding sources of the detected vibrations, and historical information regarding threat levels to the fiber optic cable. The device may perform one or more actions based on the threat level to the fiber optic cable.
H04W 4/02 - Services making use of location information
H04W 4/44 - Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
10.
RANGE AND CAPABILITY EXTENDING DEVICE FOR PERIMETER INTRUSION DETECTION
A solar powered, self-contained DFOS extender that includes a wide-angle camera and a microphone. The DFOS extender monitors the perimeter constantly or at predetermined time intervals both visually and acoustically. It analyses these signals and if an alarming event is detected (such as an animal, a human, a car, or a truck seen or heard), then the DFOS extender generates a coded vibration via its in-built acoustic modem. These coded vibrations are detected by the fiber and an event log is generated. Consequently, a trespasser or false alarms (animals) are detected before they are within detectable distance of the underground DFOS optical sensor fiber. Advantageously, our DFOS extender can be placed in critical locations along a border where an increased detection range is desired and may later be relocated/reinstalled at other locations as well.
Systems and methods for utility pole localization employing a DFOS/DAS interrogator located at one end of an optical sensor fiber remotely capture dynamic strains on the optical sensor fiber induced by acoustic events. A captured two-dimensional spatiotemporal map in an ambient noisy environment is analyzed by a trained machine learning model which then automatically detects an area in which a pole is located without requiring domain knowledge. Original DFOS/DAS signals are separated into pole regions and non-pole region time series for machine learning model training. A contrastive loss function measures similarities between low-frequency and high-frequency features. A Gaussian distribution is applied to the original signals to generate weighted labels to eliminate effects of label noise. The machine learning model fuses low-frequency and high-frequency features in the frequency domain for pole region classification. A contrastive loss is combined with cross entropy loss to measure a low-high frequency feature distance.
A computer-implemented method for ordinal prediction is provided. The method includes encoding time series data with a temporal encoder to obtain latent space representations. The method includes optimizing the temporal encoder using semi-supervised learning to distinguish different classes in the labeled space using labeled data, and augment the latent space representations using unlabeled training data, to obtain semi-supervised representations. The method further includes discarding a linear layer after the temporal encoder and fixing the temporal encoder. The method also includes training k−1 binary classifiers on top of the semi-supervised representations to obtain k−1 binary predictions. The method additionally includes identifying and correcting inconsistent ones of the k−1 binary predictions by matching the inconsistent ones to consistent ones of the k−1 binary predictions. The method further includes aggregating the k−1 binary predictions to obtain an ordinal prediction.
G06N 3/0895 - Weakly supervised learning, e.g. semi-supervised or self-supervised learning
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
Described is a novel framework, we call intent-based computing jobs assignment framework, for efficiently accommodating a clients' computing job requests in a mobile edge computing infrastructure. We define the intent-based computing job assignment problem, which jointly optimizes the virtual topology design and virtual topology mapping with the objective of minimizing the total bandwidth consumption. We use the Integer Linear Programming (ILP) technique to formulate this problem, and to facilitate the optimal solution. In addition, we employ a novel and efficient heuristic algorithm, called modified Steiner tree-based (MST-based) heuristic, which coordinately determines the virtual topology design and the virtual topology mapping. Comprehensive simulations to evaluate the performance of our solutions show that the MST-based heuristic can achieve an efficient performance that is close to the optimal performance obtained by the ILP solution.
H04L 41/0896 - Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
H04L 41/122 - Discovery or management of network topologies of virtualised topologies e.g. software-defined networks [SDN] or network function virtualisation [NFV]
14.
ROBUST SOURCE LOCALIZATION WITH JOINT TIME ARRIVAL AND VELOCITY ESTIMATION FOR CABLE CUT PREVENTION
Method for source localization for cable cut prevention using distributed fiber optic sensing (DFOS)/distributed acoustic sensing (DAS) is described that is robust/immune to underground propagation speed uncertainty. The method estimates the location of a vibration source while considering any uncertainty of vibration propagation speed and formulates the localization as an optimization problem, and both location of the sources and the propagation speed are treated as unknown. This advantageously enables our method to adapt to variances of the velocity and produce a better generalized performance with respect to environmental changes experienced in the field. Our method operates using a DFOS system and AI techniques as an integrated solution for vibration source localization along an entire optical sensor fiber cable route and process real-time DFOS data and extract features that are related to a location of a source of vibrations that may threaten optical fiber facilities.
G01V 1/37 - Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy specially adapted for seismic systems using continuous agitation of the ground
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
A flexible, rapid deployable perimeter monitoring system and method that employs distributed fiber optic sensing (DFOS) technologies and includes a deployment/operations field vehicle including an interrogator and analyzer/processor. The deployment/operations field vehicle is configured to field deploy a ruggedized fiber optic sensor cable in an arrangement that meets a specific application need, and subsequently interrogate/sense via DFOS any environmental conditions affecting the deployed fiber optic sensor cable. Such sensed conditions include mechanical vibration, acoustic, and temperature that may be advantageously sensed/evaluated/analyzed in the deployment/operations vehicle and subsequently communicated to a central location for further evaluation and/or coordination with other monitoring systems. Upon completion, the field vehicle and DFOS reconfigure a current location or redeployed to another location.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
A computer implemented method is provided for resource management of stream analytics at each individual node that includes computing a mean of output processing rate of microservices in a pipeline; and evaluating a state of each microservice of the microservices in the pipeline. The computer implemented method also includes selecting a single microservice from the pipeline for updating resources for an action that changes the state in single the microservice that is selected; and performing resource allocation update for the selected microservice. The computer implemented method may also include updating the state of the selected microservice.
A DFOS system and machine learning method that automatically localizes manholes, which forms a key step in a fiber optic cable mapping process. Our system and method utilize weakly supervised learning techniques to predict manhole locations based on ambient data captured along the fiber optic cable route. To improve any non-informative ambient data, we employ data selection and label assignment strategies and verify their effectiveness extensively in a variety of settings, including data efficiency and generalizability to different fiber optic cable routes. We describe post-processing steps that bridge the gap between classification and localization and combining results from multiple predictions.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
18.
PERSONALIZED FEDERATED LEARNING UNDER A MIXTURE OF JOINT DISTRIBUTIONS
Systems and methods for personalized federated learning. The method may include receiving at a central server local models from a plurality of clients, and aggregating a heterogeneous data distribution extracted from the local models. The method can further include processing the data distribution as a linear mixture of joint distributions to provide a global learning model, and transmitting the global learning model to the clients. The global learning model is used to update the local model.
An advance in the art is made according to aspects of the present disclosure directed to methods for earthquake sensing that employ a supervisory system of undersea fiber optic cables. Earthquakes and other environmental disturbances are detected by monitoring the polarization of interrogation light instead of its phase. More specifically, our methods monitor the transfer matrix rather than just polarization and isolate disturbance location by monitoring eigenvalues of the polarization transfer matrix. From results obtained we have demonstrated experimentally that we can monitor disturbances that affect signal polarization on a span-by-span basis using High Loss Loop Back (HLLB) paths. It is shown that by measuring the polarization rotation matrix and determining the polarization rotation angle we can identify the span where the disturbance occurred with 35 dB extinction with no limitation on the magnitude of the disturbance and the number of affected spans.
Methods and systems for training a model include distinguishing hidden states of a monitored system based on condition information. An encoder and decoder are generated for each respective hidden state using forward and backward autoencoder losses. A hybrid hidden state is determined for an input sequence based on the hidden states. The input sequence is reconstructed using the encoders and decoders and the hybrid hidden state. Parameters of the encoders and decoders are updated based on a reconstruction loss.
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
21.
HYBRID QUANTUM CRYPTOGRAPHY PROTOCOL FOR OPTICAL COMMUNICATIONS
A hybrid Quantum Key Distribution (QKD) protocol and method for secure transmission in optical communications systems and in particular long distance optical communications systems such as those in undersea environments. Our inventive method advantageously exploits fundamental security features of quantum-base encryption with the added intrinsic security of encapsulated and sealed equipment of undersea optical networks. Our method employs an intermediate node that generates an optical signal, and a pair of quantum state generators that respectively generate quantum states of the optical signal and transmit the generated quantum states to secure nodes, respectively. The secure nodes then communicate securely using a quantum key distribution (QKD) protocol.
Methods and systems for image generation include generating a latent representation of an image, modifying the latent representation of the image based on a trained attribute classifier and a specified attribute input, and decoding the modified latent representation to generate an output image that matches the specified attribute input.
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/774 - Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
23.
ADAPTIVE PERCEPTUAL QUALITY BASED CAMERA TUNING USING REINFORCEMENT LEARNING
Systems and methods are provided for dynamically tuning camera parameters in a video analytics system to optimize analytics accuracy. A camera captures a current scene, and optimal camera parameter settings are learned and identified for the current scene using a Reinforcement Learning (RL) engine. The learning includes defining a state within the RL engine as a tuple of two vectors: a first representing current camera parameter values and a second representing measured values of frames of the current scene. Quality of frames is estimated using a quality estimator, and camera parameters are adjusted based on the quality estimator and the RL engine for optimization. Effectiveness of tuning is determined using perceptual Image Quality Assessment (IQA) to quantify a quality measure. Camera parameters are adaptively tuned in real-time based on learned optimal camera parameter settings, state, quality measure, and set of actions, to optimize the analytics accuracy for video analytics tasks.
A method for generating binding peptides presented by any given Major Histocompatibility Complex (MHC) protein is presented. The method includes, given a peptide and an MHC protein pair, enabling a Reinforcement Learning (RL) agent to interact with and exploit a peptide mutation environment by repeatedly mutating the peptide and observing an observation score of the peptide, learning to form a mutation policy, via a mutation policy network, to iteratively mutate amino acids of the peptide to obtain desired presentation scores, and generating, based on the desired presentation scores, qualified peptides and binding motifs of MHC Class I proteins.
A method for generating binding peptides presented by any given Major Histocompatibility Complex (MHC) protein is presented. The method includes, given a peptide and an MHC protein pair, enabling a Reinforcement Learning (RL) agent to interact with and exploit a peptide mutation environment by repeatedly mutating the peptide and observing an observation score of the peptide, learning to form a mutation policy, via a mutation policy network, to iteratively mutate amino acids of the peptide to obtain desired presentation scores, and generating, based on the desired presentation scores, qualified peptides and binding motifs of MHC Class I proteins.
G16B 15/00 - ICT specially adapted for analysing two-dimensional or three-dimensional molecular structures, e.g. structural or functional relations or structure alignment
G16B 35/00 - ICT specially adapted for in silico combinatorial libraries of nucleic acids, proteins or peptides
In sharp contrast to the prior art, a fallen tree detection and localization method based on distributed fiber optical sensing (DFOS) technique and physics informed machine learning is described in which DFOS leverages existing fiber cables that are conventionally installed on the bottom layer of distribution lines and used to provide high-speed communications. The DFOS collects and transmits fallen tree induced vibration data along the length of the entire overhead lines, including distribution lines and transmission lines, where there is a fiber cable deployed. The developed physics-informed neural network model processes the data and localizes the fallen tree location along the lines. The location is interpreted in at least two aspects: the fallen tree location in terms of the fiber cable length; and the exact cable location (power cable or fiber cable) that the fallen tree mechanically impacts.
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
27.
VIDEO GENERATION WITH LATENT DIFFUSION PROBABILISTIC MODELS
Methods and systems for training a model include training an encoder in an unsupervised fashion based on a backward latent flow between a reference frame and a driving frame taken from a same video. A diffusion model is trained that generates a video sequence responsive to an input image and a text condition, using the trained encoder to determine a latent flow sequence and occlusion map sequence of a labeled training video.
A computer-implemented method for simulating vehicle data and improving driving scenario detection is provided. The method includes retrieving, from vehicle sensors, key parameters from real data of validation scenarios to generate corresponding scenario configurations and descriptions, transferring target scenario descriptions and validation scenario descriptions to target scenario scripts and validation scenario scripts, respectively, to create first raw simulation data pertaining to target scenario descriptions and second raw simulation data pertaining to validation scenario descriptions, training, by an adjuster network, a deep neural network model to minimize differences between the first raw simulation data and the second raw simulation data, refining the first and second raw simulation data of rare driving scenarios to generate rare driving scenario training data, and outputting the rare driving scenario training data to a display screen of a computing device to enable a user to train a scenario detector for an autonomic driving assistant system.
A computer-implemented method for training a neural network to predict object categories without manual annotation is provided. The method includes feeding training datasets including at least images and data annotations to an object detection neural network, converting, by a text prompter, the data annotations into natural text inputs, converting, by a text embedder, the natural text inputs into embeddings, minimizing objective functions during training to adjust parameters of the object detection neural network, and predicting, by the object detection neural network, objects within images and videos.
G06V 20/58 - Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
A computer-implemented method for learning disentangled representations for T-cell receptors to improve immunotherapy is provided. The method includes optionally introducing a minimal number of mutations to a T-cell receptor (TCR) sequence to enable the TCR sequence to bind to a peptide, using a disentangled Wasserstein autoencoder to separate an embedding space of the TCR sequence into functional embeddings and structural embeddings, feeding the functional embeddings and the structural embeddings to a long short-term memory (LSTM) or transformer decoder, using an auxiliary classifier to predict a probability of a positive binding label from the functional embeddings and the peptide, and generating new TCR sequences with enhanced binding affinity for immunotherapy to target a particular virus or tumor.
Methods and systems for training a language model include retrieving a knowledge sentence, related to an input sentence, from a knowledge base. The input sentence, the knowledge sentence, and a prompt are encoded into an intermediate representation. The intermediate representation is decoded to generate a named entity from the input sentence that is of a type specified by the prompt. A language model is fine-tuned based on the named entity.
Methods and systems for training a model include pre-training a backbone model with a pre-training decoder, using an unlabeled dataset with multiple distinct sensor data modalities that derive from different sensor types. The backbone model is fine-tuned with an output decoder after pre-training, using a labeled dataset with the multiple modalities.
G06V 20/56 - Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
G06V 10/774 - Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
G06V 10/80 - Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
33.
BINDING PEPTIDE GENERATION FOR MHC CLASS I PROTEINS WITH DEEP REINFORCEMENT LEARNING
A method for generating binding peptides presented by any given Major Histocompatibility Complex (MHC) protein is presented. The method includes, given a peptide and an MHC protein pair, enabling a Reinforcement Learning (RL) agent to interact with and exploit a peptide mutation environment by repeatedly mutating the peptide and observing an observation score of the peptide, learning to form a mutation policy, via a mutation policy network, to iteratively mutate amino acids of the peptide to obtain desired presentation scores, and generating, based on the desired presentation scores, qualified peptides and binding motifs of MHC Class I proteins.
G16B 15/00 - ICT specially adapted for analysing two-dimensional or three-dimensional molecular structures, e.g. structural or functional relations or structure alignment
G16B 35/00 - ICT specially adapted for in silico combinatorial libraries of nucleic acids, proteins or peptides
A system for binding peptide search for immunotherapy is presented. The system includes employing a deep neural network to predict a peptide presentation given Major Histocompatibility Complex allele sequences and peptide sequences, training a Variational Autoencoder (VAE) to reconstruct peptides by converting the peptide sequences into continuous embedding vectors, running a Monte Carlo Tree Search to generate a first set of positive peptide vaccine candidates, running a Bayesian Optimization search with the trained VAE and a Backpropagation search with the trained VAE to generate a second set of positive peptide vaccine candidates, using a sampling from a Position Weight Matrix (sPWM) to generate a third set of positive peptide vaccine candidates, screening and merging the first, second, and third sets of positive peptide vaccine candidates, and outputting qualified peptides for immunotherapy from the screened and merged sets of positive peptide vaccine candidates.
G16B 40/00 - ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
A computer-implemented method for detecting objects within an advanced driver assistance system (ADAS) is provided. The method includes obtaining road scene datasets from a plurality of cameras, including at least road scene images and road scene data annotations, to be provided to an object detection neural network communicating with an open-vocabulary detector of a vehicle, converting, by a text prompter, the road scene data annotations into natural text inputs, converting, by a text embedder, the natural text inputs into embeddings, minimizing objective functions during training to adjust parameters of the object detection neural network, and detecting, by the object detection neural network, objects within the road scene datasets to provide alerts or notifications to a driver of the vehicle pertaining to the detected objects.
G06V 20/56 - Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/774 - Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
Methods and systems for training a model include determining class prototypes of time series samples from a training dataset. A task corresponding to the time series samples is encoded using the class prototypes and a task-level configuration. A likelihood value is determined based on outputs of a time series density model, a task-class distance from a task embedding model, and a task density model. Parameters of the time series density model, the task embedding model, and the task density model are adjusted responsive to the likelihood value.
G06F 18/2413 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
G06F 18/2415 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
H04B 10/69 - Electrical arrangements in the receiver
A system for binding peptide search for immunotherapy is presented. The system includes employing a deep neural network to predict a peptide presentation given Major Histocompatibility Complex allele sequences and peptide sequences, training a Variational Autoencoder (VAE) to reconstruct peptides by converting the peptide sequences into continuous embedding vectors, running a Monte Carlo Tree Search to generate a first set of positive peptide vaccine candidates, running a Bayesian Optimization search with the trained VAE and a Backpropagation search with the trained VAE to generate a second set of positive peptide vaccine candidates, using a sampling from a Position Weight Matrix (sPWM) to generate a third set of positive peptide vaccine candidates, screening and merging the first, second, and third sets of positive peptide vaccine candidates, and outputting qualified peptides for immunotherapy from the screened and merged sets of positive peptide vaccine candidates.
G16B 40/00 - ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
A system for binding peptide search for immunotherapy is presented. The system includes employing a deep neural network to predict a peptide presentation given Major Histocompatibility Complex allele sequences and peptide sequences, training a Variational Autoencoder (VAE) to reconstruct peptides by converting the peptide sequences into continuous embedding vectors, running a Monte Carlo Tree Search to generate a first set of positive peptide vaccine candidates, running a Bayesian Optimization search with the trained VAE and a Backpropagation search with the trained VAE to generate a second set of positive peptide vaccine candidates, using a sampling from a Position Weight Matrix (sPWM) to generate a third set of positive peptide vaccine candidates, screening and merging the first, second, and third sets of positive peptide vaccine candidates, and outputting qualified peptides for immunotherapy from the screened and merged sets of positive peptide vaccine candidates.
G16B 40/00 - ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
A computer-implemented method for employing a time-series-to-text generation model to generate accurate description texts is provided. The method includes passing time series data through a time series encoder and a multilayer perceptron (MLP) classifier to obtain predicted concept labels, converting the predicted concept labels, by a serializer, to a text token sequence by concatenating an aspect term and an option term of every aspect, inputting the text token sequence into a pretrained language model including a bidirectional encoder and an autoregressive decoder, and using adapter layers to fine-tune the pretrained language model to generate description texts.
A computer-implemented method for identifying root cause failure and fault events is provided. The method includes detecting a trigger point, converting, via an encoder, previous system state data, new batch data in a next system state, and a causal graph to system state-invariant embeddings and system state-dependent embeddings, generating a learned causal graph, via a graph generation layer, by integrating state-invariant and state-dependent information, and predicting, by a prediction layer, future time-series data on the learned causal graph.
A computer-implemented method for employing a graph-based adaptive domain generation framework is provided. The method includes, in a training phase, performing domain prototypical network training on source domains, constructing an autoencoding domain relation graph by applying a graph autoencoder to produce domain node embeddings, and performing, via a domain-adaptive classifier, domain-adaptive classifier training to make an informed decision. The method further includes, in a testing phase, given testing samples from a new source domain, computing a prototype by using a pretrained domain prototypical network, inferring node embedding, and making a prediction by the domain-adaptive classifier based on the domain node embeddings.
Methods and systems for training a model include performing skill discovery, using a set of demonstrations that includes known-good demonstrations and noisy demonstrations, to generate a set of skills. A unidirectional skill embedding model is trained in a first training while parameters of a skill matching model and low-level policies that relate skills to actions are held constant. The unidirectional skill embedding model, the skill matching model, and the low-level policies are trained together in an end-to-end fashion in a second training.
A computer-implemented method for employing a graph-based log anomaly detection framework to detect relational anomalies in system logs is provided. The method includes collecting log events from systems or applications or sensors or instruments, constructing dynamic graphs to describe relationships among the log events and log fields by using a sliding window with a fixed time interval to snapshot a batch of the log events, capturing sequential patterns by employing temporal-attentive transformers to learn temporal dependencies within the sequential patterns, and detecting anomalous patterns in the log events based on relationships between the log events and temporal context determined from the temporal-attentive transformers.
H04L 41/16 - Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
44.
DISENTANGLED GRAPH LEARNING FOR INCREMENTAL CAUSAL DISCOVERY AND ROOT CAUSE ANALYSIS
A computer-implemented method for locating root causes is provided. The method includes detecting a trigger point from entity metrics data and key performance indicator (KPI) data, generating a learned causal graph by fusing a state-invariant causal graph with a state-dependent causal graph, and locating the root causes by employing a random walk-based technique to estimate a probability score for each of the entity metrics data by starting from a KPI node.
A computer-implemented method for counting and extracting opinions in product reviews is provided. The method includes inputting a hypothesis opinion, a product name, and product reviews relating to a product, applying a decontextualization component to the product reviews by using the product name, applying the decontextualization component to the hypothesis opinion by using the product name, applying an entailment model to classify each sentence of the decontextualized product reviews against the decontextualized hypothesis opinion, and outputting one or more sentences classified as entailing the hypothesis opinion and a count of corresponding reviews.
Methods and systems for transmission over a heterogeneous network include determining a path through a first network, including collecting quality of service (QoS) parameters for network devices in the path. The QoS parameters of the network devices are configured to provide a predetermined QoS assurance across the path. A network flow is transmitted from a network slice of a second network through the first network, along the path.
Methods and systems for object localization include identifying associations between measurements taken from radar sensors. A shared coordinate system for the radar sensors is determined based the identified associations, including identifying translations and rotations between local coordinate systems of the radar sensors. A position of an object in the shared coordinate systems is determined, based on measurements of the object by the radar sensors. An action is performed responsive to the determined position of the object.
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/87 - Combinations of radar systems, e.g. primary radar and secondary radar
G01S 13/42 - Simultaneous measurement of distance and other coordinates
Methods and systems for video processing include enriching an input video feature from an input video frame set using a meta-action bank video sub-actions to generate enriched features. Reinforced image representation is performed using reinforcement learning to compare support image frames and query image frames and determine an importance of the input video frame. A classification is performed on the input video frame based on the importance and the enriched features to generate a label. An action is performed responsive to the generated label.
Methods and systems for video processing include extracting flow features and appearance features from frames of a video stream. The flow features are processed using a flow model that is trained on a first set of training data. An output of the flow model is processed using a sub-network that is trained on the first set of training data and a second set of domain-specific training data to generate a flow parameter. The appearance features are processed using an appearance model that is trained on the first set of training data and that further processes the appearance features using the flow parameter, to classify the frames of the video stream. An action is performed responsive to the classified frames.
Systems and methods for performing the dynamic anomaly localization of utility pole aerial/suspended/supported wires/cables by distributed fiber optic sensing. In sharp contrast to the prior art, our inventive systems and methods according to aspects of the present disclosure advantageously identify a “location region” on a utility pole supporting an affected wire/cable, thereby permitting the identification and reporting of service personnel that are uniquely responsible for responding to such anomalous condition(s).
H02G 1/02 - Methods or apparatus specially adapted for installing, maintaining, repairing, or dismantling electric cables or lines for overhead lines or cables
G01S 19/01 - Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
51.
TRIGGER POINT DETECTION FOR ONLINE ROOT CAUSE ANALYSIS AND SYSTEM FAULT DIAGNOSIS
A computer-implemented method for detecting trigger points to identify root cause failure and fault events is provided. The method includes collecting, by a monitoring agent, entity metrics data and system key performance indicator (KPI) data, integrating the entity metrics data and the KPI data, constructing an initial system state space, detecting system state changes by calculating a distance between current batch data and an initial state, and dividing a system status into different states.
A method for learning a self-explainable imitator by discovering causal relationships between states and actions is presented. The method includes obtaining, via an acquisition component, demonstrations of a target task from experts for training a model to generate a learned policy, training the model, via a learning component, the learning component computing actions to be taken with respect to states, generating, via a dynamic causal discovery component, dynamic causal graphs for each environment state, encoding, via a causal encoding component, discovered causal relationships by updating state variable embeddings, and outputting, via an output component, the learned policy including trajectories similar to the demonstrations from the experts.
A method for learning a self-explainable imitator by discovering causal relationships between states and actions is presented. The method includes obtaining, via an acquisition component, demonstrations of a target task from experts for training a model to generate a learned policy, training the model, via a learning component, the learning component computing actions to be taken with respect to states, generating, via a dynamic causal discovery component, dynamic causal graphs for each environment state, encoding, via a causal encoding component, discovered causal relationships by updating state variable embeddings, and outputting, via an output component, the learned policy including trajectories similar to the demonstrations from the experts.
A method for acquiring skills through imitation learning by employing a meta imitation learning framework with structured skill discovery (MILD) is presented. The method includes learning behaviors or tasks, by an agent, from demonstrations: by learning to decompose the demonstrations into segments, via a segmentation component, the segments corresponding to skills that are transferrable across different tasks, learning relationships between the skills that are transferrable across the different tasks, employing, via a graph generator, a graph neural network for learning implicit structures of the skills from the demonstrations to define structured skills, and generating policies from the structured skills to allow the agent to acquire the structured skills for application to one or more target tasks.
Methods and systems for temporal action localization include processing a video stream to identify an action and a start time and a stop time for the action using a neural network model that separately processes information of appearance and motion modalities from the video stream using transformer branches that include a self-attention and a cross-attention between the appearance and motion modalities. An action is performed responsive to the identified action.
G06V 10/77 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
A method for learning a self-explainable imitator by discovering causal relationships between states and actions is presented. The method includes obtaining, via an acquisition component, demonstrations of a target task from experts for training a model to generate a learned policy, training the model, via a learning component, the learning component computing actions to be taken with respect to states, generating, via a dynamic causal discovery component, dynamic causal graphs for each environment state, encoding, via a causal encoding component, discovered causal relationships by updating state variable embeddings, and outputting, via an output component, the learned policy including trajectories similar to the demonstrations from the experts.
G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
57.
META IMITATION LEARNING WITH STRUCTURED SKILL DISCOVERY
A method for acquiring skills through imitation learning by employing a meta imitation learning framework with structured skill discovery (MILD) is presented. The method includes learning behaviors or tasks, by an agent, from demonstrations: by learning to decompose the demonstrations into segments, via a segmentation component, the segments corresponding to skills that are transferrable across different tasks, learning relationships between the skills that are transferrable across the different tasks, employing, via a graph generator, a graph neural network for learning implicit structures of the skills from the demonstrations to define structured skills, and generating policies from the structured skills to allow the agent to acquire the structured skills for application to one or more target tasks.
A method interprets a convolutional sequence model. The method converts an input data sequence having input segments into output features. The method clusters the input segments into clusters using respective resolution-controllable class prototypes allocated to each of classes. Each respective class prototype includes a respective output feature subset characterizing a respective associated class. The method calculates, using the clusters, similarity scores that indicate a similarity of an output feature to a respective class prototypes responsive to distances between the output feature and the respective class prototypes. The method concatenates the similarity scores to obtain a similarity vector. The method performs a prediction and prediction support operation that provides a value of prediction and an interpretation for the value responsive to the input segments and similarity vector. The interpretation for the value of prediction is provided using only non-negative weights and lacking a weight bias in the fully connected layer.
Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.
A method for performing contrastive learning for graph tasks and datasets by employing an information-aware graph contrastive learning framework is presented. The method includes obtaining two semantically similar views of a graph coupled with a label for training by employing a view augmentation component, feeding the two semantically similar views into respective encoder networks to extract latent representations preserving both structure and attribute information in the two views, optimizing a contrastive loss based on a contrastive mode by maximizing feature consistency between the latent representations, training a neural network with the optimized contrastive loss, and predicting a new graph label or a new node label in the graph.
Systems and methods are provided for increasing accuracy of video analytics tasks in real-time by acquiring a video using video cameras, and identifying fluctuations in the accuracy of video analytics applications across consecutive frames of the video. The identified fluctuations are quantified based on an average relative difference of true-positive detection counts across consecutive frames. Fluctuations in accuracy are reduced by applying transfer learning to a deep learning model initially trained using images, and retraining the deep learning model using video frames. A quality of object detections is determined based on an amount of track-ids assigned by a tracker across different video frames. Optimization of the reduction of fluctuations includes iteratively repeating the identifying, the quantifying, the reducing, and the determining the quality of object detections until a threshold is reached. Model predictions for each frame in the video are generated using the retrained deep learning model.
A method for acquiring skills through imitation learning by employing a meta imitation learning framework with structured skill discovery (MILD) is presented. The method includes learning behaviors or tasks, by an agent, from demonstrations: by learning to decompose the demonstrations into segments, via a segmentation component, the segments corresponding to skills that are transferrable across different tasks, learning relationships between the skills that are transferrable across the different tasks, employing, via a graph generator, a graph neural network for learning implicit structures of the skills from the demonstrations to define structured skills, and generating policies from the structured skills to allow the agent to acquire the structured skills for application to one or more target tasks.
Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.
Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.
A method for performing contrastive learning for graph tasks and datasets by employing an information-aware graph contrastive learning framework is presented. The method includes obtaining two semantically similar views of a graph coupled with a label for training by employing a view augmentation component, feeding the two semantically similar views into respective encoder networks to extract latent representations preserving both structure and attribute information in the two views, optimizing a contrastive loss based on a contrastive mode by maximizing feature consistency between the latent representations, training a neural network with the optimized contrastive loss, and predicting a new graph label or a new node label in the graph.
A method for performing contrastive learning for graph tasks and datasets by employing an information-aware graph contrastive learning framework is presented. The method includes obtaining two semantically similar views of a graph coupled with a label for training by employing a view augmentation component, feeding the two semantically similar views into respective encoder networks to extract latent representations preserving both structure and attribute information in the two views, optimizing a contrastive loss based on a contrastive mode by maximizing feature consistency between the latent representations, training a neural network with the optimized contrastive loss, and predicting a new graph label or a new node label in the graph.
A method interprets a convolutional sequence model. The method converts an input data sequence having input segments into output features. The method clusters the input segments into clusters using respective resolution-controllable class prototypes allocated to each of classes. Each respective class prototype includes a respective output feature subset characterizing a respective associated class. The method calculates, using the clusters, similarity scores that indicate a similarity of an output feature to a respective class prototypes responsive to distances between the output feature and the respective class prototypes. The method concatenates the similarity scores to obtain a similarity vector. The method performs a prediction and prediction support operation that provides a value of prediction and an interpretation for the value responsive to the input segments and similarity vector. The interpretation for the value of prediction is provided using only non-negative weights and lacking a weight bias in the fully connected layer.
A method interprets a convolutional sequence model. The method converts an input data sequence having input segments into output features. The method clusters the input segments into clusters using respective resolution-controllable class prototypes allocated to each of classes. Each respective class prototype includes a respective output feature subset characterizing a respective associated class. The method calculates, using the clusters, similarity scores that indicate a similarity of an output feature to a respective class prototypes responsive to distances between the output feature and the respective class prototypes. The method concatenates the similarity scores to obtain a similarity vector. The method performs a prediction and prediction support operation that provides a value of prediction and an interpretation for the value responsive to the input segments and similarity vector. The interpretation for the value of prediction is provided using only non-negative weights and lacking a weight bias in the fully connected layer.
A computer-implemented method is provided for generating new binding peptides to Major Histocompatibility Complex (MHC) proteins. The method includes training, by a processor device, a Generative Adversarial Network GAN having a generator and a discriminator only on a set of binding peptide sequences given training data comprising the set of binding peptide sequences and a set of non-binding peptide sequences. A GAN training objective includes the discriminator being iteratively updated to distinguish generated peptide sequences from sampled binding peptide sequences as fake or real and the generator being iteratively updated to fool the discriminator. The training includes optimizing the GAN training objective while learning two projection vectors for a binding class with two cross-entropy losses. A first loss discriminating binding peptide sequences in the training data from non-binding peptide sequences in the training data. A second loss discriminating generated binding peptide sequences from non-binding peptide sequences in the training data.
A computer-implemented method is provided for generating new binding peptides to Major Histocompatibility Complex (MHC) proteins. The method includes training, by a processor device, a Generative Adversarial Network GAN having a generator and a discriminator only on a set of binding peptide sequences given training data comprising the set of binding peptide sequences and a set of non-binding peptide sequences. A GAN training objective includes the discriminator being iteratively updated to distinguish generated peptide sequences from sampled binding peptide sequences as fake or real and the generator being iteratively updated to fool the discriminator. The training includes optimizing the GAN training objective while learning two projection vectors for a binding class with two cross-entropy losses. A first loss discriminating binding peptide sequences in the training data from non-binding peptide sequences in the training data. A second loss discriminating generated binding peptide sequences from non-binding peptide sequences in the training data.
A computer-implemented method is provided for generating new binding peptides to Major Histocompatibility Complex (MHC) proteins. The method includes training, by a processor device, a Generative Adversarial Network GAN having a generator and a discriminator only on a set of binding peptide sequences given training data comprising the set of binding peptide sequences and a set of non-binding peptide sequences. A GAN training objective includes the discriminator being iteratively updated to distinguish generated peptide sequences from sampled binding peptide sequences as fake or real and the generator being iteratively updated to fool the discriminator. The training includes optimizing the GAN training objective while learning two projection vectors for a binding class with two cross-entropy losses. A first loss discriminating binding peptide sequences in the training data from non-binding peptide sequences in the training data. A second loss discriminating generated binding peptide sequences from non-binding peptide sequences in the training data.
Systems and methods for predicting an occurrence of a medical event for a patient using a trained neural network. Historical patient data is preprocessed to generate normalized training samples, and the normalized training samples are sent to a personalized deep convolutional neural network for model pretraining and updating of model parameters. The pretrained model is stored in a remote server for utilization by a local machine for personalization during a preparation time period for a medical treatment. A normalized finetuning set is generated as output, and the model parameters are iteratively finetuned. A personal prediction score for future medical events is generated, and an operation of a medical treatment device is controlled responsive to the prediction score.
G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Methods and systems for responding to changing conditions include training a model, using a processor, using trajectories that resulted in a positive outcome and trajectories that resulted in a negative outcome. Training is performed using an adversarial discriminator to train the model to generate trajectories that are similar to historical trajectories that resulted in a positive outcome, and using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical trajectories that resulted in a negative outcome. A dynamic response regime is generated using the trained model and environment information. A response to changing environment conditions is performed in accordance with the dynamic response regime.
G06N 3/084 - Backpropagation, e.g. using gradient descent
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
A computer-implemented method for model building is provided. The method includes receiving a training set of medical records and model hyperparameters. The method further includes initializing an encoder as a Dual-Channel Combiner Network (DCNN) and initialize distribution related parameters. The method also includes performing, by a hardware processor, a forward computation to (1) the DCNN to obtain the embeddings of the medical records, and (2) the distribution related parameters to obtain class probabilities. The method additionally includes checking by a convergence evaluator if the iterative optimization has converged. The method further includes performing model personalization responsive to model convergence by encoding the support data of a new patient and using the embeddings and event subtype labels to train a personalized classifier.
A computer-implemented method for model building is provided. The method includes receiving a training set of medical records and model hyperparameters. The method further includes initializing an encoder as a Dual-Channel Combiner Network (DCNN) and initialize distribution related parameters. The method also includes performing, by a hardware processor, a forward computation to (1) the DCNN to obtain the embeddings of the medical records, and (2) the distribution related parameters to obtain class probabilities. The method additionally includes checking by a convergence evaluator if the iterative optimization has converged. The method further includes performing model personalization responsive to model convergence by encoding the support data of a new patient and using the embeddings and event subtype labels to train a personalized classifier.
A DFOS system and method providing detection and localization of vehicle emergency stops which employs DFOS and machine learning techniques as an integrated solution for automatic, real-time, detection and localization of vehicle emergency stop events. Real time-location data from a buried optical sensing fiber located along a roadway is used to derive continuous vehicle trajectories while providing a wide coverage area for more accurate assessments. AI techniques are employed which track vehicles' speed and acceleration, locate vehicle deceleration events, and localize emergency stop events. Danger assessment is determined by analyzing a vehicle's anticipated and reproducible trajectory after an emergency breaking event—while ignoring stop-and-go events as low-risk events—and discovering stop-no-go events exhibiting large deceleration as high-risk events.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
Systems and methods for predicting an occurrence of a medical event for a patient using a trained neural network. Historical patient data is preprocessed to generate normalized training samples, and the normalized training samples are sent to a personalized deep convolutional neural network for model pretraining and updating of model parameters. The pretrained model is stored in a remote server for utilization by a local machine for personalization during a preparation time period for a medical treatment. A normalized finetuning set is generated as output, and the model parameters are iteratively finetuned. A personal prediction score for future medical events is generated, and an operation of a medical treatment device is controlled responsive to the prediction score.
G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Systems and methods for predicting an occurrence of a medical event for a patient using a trained neural network. Historical patient data is preprocessed to generate normalized training samples, and the normalized training samples are sent to a personalized deep convolutional neural network for model pretraining and updating of model parameters. The pretrained model is stored in a remote server for utilization by a local machine for personalization during a preparation time period for a medical treatment. A normalized finetuning set is generated as output, and the model parameters are iteratively finetuned. A personal prediction score for future medical events is generated, and an operation of a medical treatment device is controlled responsive to the prediction score.
G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
A computer-implemented method for model building is provided. The method includes receiving a training set of medical records and model hyperparameters. The method further includes initializing an encoder as a Dual-Channel Combiner Network (DCNN) and initialize distribution related parameters. The method also includes performing, by a hardware processor, a forward computation to (1) the DCNN to obtain the embeddings of the medical records, and (2) the distribution related parameters to obtain class probabilities. The method additionally includes checking by a convergence evaluator if the iterative optimization has converged. The method further includes performing model personalization responsive to model convergence by encoding the support data of a new patient and using the embeddings and event subtype labels to train a personalized classifier.
A method for time synchronization using distributed fiber optic sensing (DFOS) that employs several trusted time beacons that are attached to the DFOS sensing fiber which in turn is connected to the DFOS interrogator. The beacons transmit their signal via two different mediums, (1) wirelessly to sensor nodes in the coverage area, and (2) through vibrations on fiber to the DFOS/DAS system located at a trusted area such as the central office. Wireless broadcast to nearby sensors includes a timestamp and beacon ID. All the sensors in the field use one of the beacons in their vicinity (the one with the strongest signal) as their time reference and send the data back with the corresponding beacon index.
H04B 10/071 - Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using a reflected signal, e.g. using optical time domain reflectometers [OTDR]
H04B 10/2575 - Radio-over-fibre, e.g. radio frequency signal modulated onto an optical carrier
81.
FLEXIBLE AND EFFICIENT COMMUNICATION IN MICROSERVICES-BASED STREAM ANALYTICS PIPELINE
A pull-based communication method for microservices-based real-time streaming video analytics pipelines is provided. The method includes receiving a plurality of frames from a plurality of cameras, each camera including a camera sidecar, arranging a plurality of detectors in layers such that a first detector layer includes detectors with detector sidecars and detector business logic, and the second detector layer includes detectors with only sidecars, arranging a plurality of extractors in layers such that a first extractor layer includes extractors with extractor sidecars and extractor business logic, and the second extractor layer includes extractors with only sidecars, and enabling a mesh controller, during registration, to selectively assign inputs to one or more of the detector sidecars of the first detector layer and one or more of the extractor sidecars of the first extractor layer to pull data items for processing.
Methods and systems for event detection include training a joint neural network model with respective neural networks for audio data and video data relating to a same scene. The joint neural network model is configured to output a belief value, a disbelief value, and an uncertainty value. It is determined that an event has occurred based on the belief value, the disbelief value, and the uncertainty value.
A machine learning (ML)/artificial intelligence (AI) based distributed fiber optic sensing (DFOS) system and method providing detection and localization of gunshot events. In addition to the ML/AI DFOS, a signal processing pipeline that compresses an audible distributed acoustic sensing (DAS) waveform data into a small set of features that protects privacy of individuals while preserving the utility of acoustic events to detect the gunshot events and discriminate same from other events. A data-driven deep learning approach automatically predicts acoustic event types with higher accuracy that realized by prior art methods.
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
F41A 19/01 - Counting means indicating the number of shots fired
84.
PERSONALIZED FEDERATED LEARNING VIA HETEROGENEOUS MODULAR NETWORKS
A computer-implemented method for personalizing heterogeneous clients is provided. The method includes initializing a federated modular network including a plurality of clients communicating with a server, maintaining, within the server, a heterogenous module pool having sub-blocks and a routing hypernetwork, partitioning the plurality of clients by modeling a joint distribution of each client into clusters, enabling each client to make a decision in each update to assemble a personalized model by selecting a combination of sub-blocks from the heterogenous module pool, and generating, by the routing hypernetwork, the decision for each client.
H04L 41/16 - Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
H04L 41/142 - Network analysis or design using statistical or mathematical methods
85.
Tensorized LSTM with Adaptive Shared Memory for Learning Trends in Multivariate Time Series
A method for executing a multi-task deep learning model for learning trends in multivariate time series is presented. The method includes collecting multi-variate time series data from a plurality of sensors, jointly learning both local and global contextual features for predicting a trend of the multivariate time series by employing a tensorized long short-term memory (LSTM) with adaptive shared memory (TLASM) to learn historical dependency of historical trends, and employing a multi-task one-dimensional convolutional neural network (1dCNN) to extract salient features from local raw time series data to model a short-term dependency between local time series data and subsequent trends.
Methods and systems for reserving resources include determining a state of a distributed computing system based on resource needs of an application that is executed on the distributed computing system and system resource constraints. An action is determined using the state of the distributed computing system as an input to a trained reinforcement learning model. A resource request is issued for the application to reserve resources based on the action.
A method for detecting pod and node candidates from cloud computing systems representing potential root causes of failure or fault activities is presented. The method includes collecting, by a monitoring agent, multi-modality data including key performance indicator (KPI) data, metrics data, and log data, employing a feature extractor and representation learner to convert the log data to time series data, applying a metric prioritizer based on extreme value theory to prioritize metrics for root cause analysis and learn an importance of different metrics, ranking root causes of failure or fault activities by using a hierarchical graph neural network, and generating one or more root cause reports outlining the potential root causes of failure or fault activities.
A distributed fiber optic sensing (DFOS) system including a smart-mat that: 1) identifies indoor locations of moving persons/objects; 2) provides a 2D visual mapping; and 3) covers any blind spots with supplemental technologies including LiDAR, RF radar, etc. The DFOS system with smart-mat may be deployed virtually anywhere indoors and may even be constructed to replace carpeting. When our inventive DFOS system and smart-mat is deployed throughout an entire building, building safety and security is greatly improved by providing up-to-date localization information of building occupants while eliminating blind zones and reducing maintenance costs.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
89.
AUTOMATIC FIBER LOSS DETECTION USING COHERENT OTDR
Computer vision based, wide-area snow/water level estimation methods using disparity maps. In one embodiment our method provides rich depth-information using a stereo camera and image processing. Scene images at normal and snow/rain weather conditions are obtained by a double-lens stereo camera and a disparity map is generated from the scene images at left and right lenses using a self-supervised deep convolutional network. In another embodiment, our method uses a single point snow/water level sensor, a stationary monocular camera to measure snow/water levels covering a wide area.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
G01M 11/00 - Testing of optical apparatus; Testing structures by optical methods not otherwise provided for
90.
Adversarial Cooperative Imitation Learning for Dynamic Treatment
Methods and systems for responding to changing conditions include training a model, using a processor, using trajectories that resulted in a positive outcome and trajectories that resulted in a negative outcome. Training is performed using an adversarial discriminator to train the model to generate trajectories that are similar to historical trajectories that resulted in a positive outcome, and using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical trajectories that resulted in a negative outcome. A dynamic response regime is generated using the trained model and environment information. A response to changing environment conditions is performed in accordance with the dynamic response regime.
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
91.
Adversarial Cooperative Imitation Learning for Dynamic Treatment
Methods and systems for responding to changing conditions include training a model, using a processor, using trajectories that resulted in a positive outcome and trajectories that resulted in a negative outcome. Training is performed using an adversarial discriminator to train the model to generate trajectories that are similar to historical trajectories that resulted in a positive outcome, and using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical trajectories that resulted in a negative outcome. A dynamic response regime is generated using the trained model and environment information. A response to changing environment conditions is performed in accordance with the dynamic response regime.
G06N 3/084 - Backpropagation, e.g. using gradient descent
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
A method for detecting an origin of a computer attack given a detection point based on multi-modality data is presented. The method includes monitoring a plurality of hosts in different enterprise system entities to audit log data and metrics data, generating causal dependency graphs to learn statistical causal relationships between the different enterprise system entities based on the log data and the metrics data, detecting a computer attack by pinpointing attack detection points, backtracking from the attack detection points by employing the causal dependency graphs to locate an origin of the computer attack, and analyzing computer attack data resulting from the backtracking to prevent present and future computer attacks.
G06F 21/55 - Detecting local intrusion or implementing counter-measures
G06F 21/57 - Certifying or maintaining trusted computer platforms, e.g. secure boots or power-downs, version controls, system software checks, secure updates or assessing vulnerabilities
A method for employing root cause analysis is presented. The method includes embedding, by an embedding layer, a sequence of events into a low-dimension space, employing a feature extractor and representation learner to convert log data from the sequence of events to time series data, the feature extractor including an auto-encoder model and a language model, and detecting root causes of failure or fault activities from the time series data.
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
G06N 7/01 - Probabilistic graphical models, e.g. probabilistic networks
94.
LOW-COST HIGH PRECISION BAROMETRIC PRESSURE MEASUREMENT SYSTEM
Disclosed are systems and methods to determine barometric pressure 1) using multiple low-cost pressure sensors located at known heights instead of a single high-cost sensor; 2) determines an actual pressure value—not by averaging multiple sensors but rather optimizing an expected error in each individual one of them and utilize their known sensor heights thereby defining a new error function; 3) our approach is scalable, i.e. the number of sensors can be increased, and multiple sensors can be grouped together into smaller cells such that each group of cell can be corrected separately, and can even be corrected among themselves. Finally, our systems and methods according to the present disclosure can advantageously be integrated with a distributed fiber optic sensing (DFOS) system via acoustic modems thereby providing extremely wide-area external, or interior buildings pressure readings.
G01C 5/06 - Measuring height; Measuring distances transverse to line of sight; Levelling between separated points; Surveyors' levels by using barometric means
G01K 11/3206 - Measuring temperature based on physical or chemical changes not covered by group , , , or using changes in transmittance, scattering or luminescence in optical fibres at discrete locations in the fibre, e.g. using Bragg scattering
95.
Telecom Cable Tension Screening Technique Based on Wave Propagation and Distributed Acoustic Sensing
A radio-controlled, two-way acoustic modem for operating with a distributed fiber optic sensing (DFOS) system including circuitry that receives radio signals including configuration information, configures the modem to operate according to the configuration information, and generate acoustic signals that are detected by the DFOS system. The acoustic modem includes one or more sensors that detect environmental information that is encoded in the acoustic signals for further reception by the DFOS system. The received configuration information may change the operating times, sensors or other operating aspects of the modem as desired an such information may be transmitted from a fixed location or a mobile vehicle.
G01D 5/353 - Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using optical means, i.e. using infrared, visible or ultraviolet light with attenuation or whole or partial obturation of beams of light the beams of light being detected by photocells influencing the transmission properties of an optical fibre
G06T 7/70 - Determining position or orientation of objects or cameras
G01H 9/00 - Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
G01L 1/24 - Measuring force or stress, in general by measuring variations of optical properties of material when it is stressed, e.g. by photoelastic stress analysis
96.
DYNAMIC INTENT-BASED NETWORK COMPUTING JOB ASSIGNMENT USING REINFORCEMENT LEARNING
An advance in the art is made according to aspects of the present disclosure directed to a method that determines virtual topology design and resource allocation for dynamic intent-based computing jobs in a mobile edge computing infrastructure when client requests are dynamic. Our method according to aspects of the present disclosure is an unsupervised machine learning approach, so that there is no need for manual labeling or pre-processing in advance, while a training process and decision making is performed online. In sharp contrast to the prior art, our method according to aspects of the present disclosure utilizes reinforcement learning techniques to make an efficient assignment in which two neural networks—a policy neural network and a value neural network—are used interactively to achieve the assignment. A training process is performed through a batch (or group) processing style in an online manner.
Computer vision based, wide-area snow/water level estimation methods using disparity maps. In one embodiment our method provides rich depth-information using a stereo camera and image processing. Scene images at normal and snow/rain weather conditions are obtained by a double-lens stereo camera and a disparity map is generated from the scene images at left and right lenses using a self-supervised deep convolutional network. In another embodiment, our method uses a single point snow/water level sensor, a stationary monocular camera to measure snow/water levels covering a wide area.
Methods and systems for peptide generation include training a peptide mutation policy neural network using reinforcement learning that includes a peptide presentation score as a reward. New peptides are generated using the peptide mutation policy. A binding motif of a major histocompatibility complex is calculated using the new peptides. Library peptides are screened in accordance with the binding motif.
A distributed acoustic sensing (DAS) system that collects many frequency bands within the system (in particular, analog to digital converter, or ADC) bandwidth. The bands are divided into multiple groups, and within each group, the inter-band interference from band i to band j is at least l1 dB lower than the power of band j. For different groups, the inter-band interference is at least l2 dB lower than the power of band j, that l2≥l1+Cl, where Cl is a value related to round-trip fiber loss. The bands belonging to the same group are cascaded one after another, with little guard time in between, or packed back-to-back. This arrangement advantageously increases the overall signal power, resulting in an improved SNR.
A distributed fiber optic sensing (DFOS) / distributed acoustic sensing (DAS) system and method employing a fiber optic sensor cable that collects vibrational data of individual utility poles suspending the fiber optic sensor cable and stores the vibrational data in a central office (CO). Machine learning (ML) models are developed, trained, and utilized to analyze vibrational features of the utility poles and determine their integrity. Additionally, DFOS / DAS systems and methods according to the present disclosure determine the location(s) of fiber coils that exist along a length of a fiber optic sensor cable.
patents
