Tata Consultancy Services Limited

Arijit Ukil

Arpan Pal

Soma Bandyopadhyay

Chetanya Puri

Rituraj Singh

C A Murthy

Rohan Banerjee

Kayapanda Muthana Mandana

Ishan Sahu

Arijit Mukherjee

Swarnava Dey

Avik Ghose

Sachin Premsukh Lodha

Puneet Kumar Agarwal

Chirabrata Bhaumik

Tanushyam Chattopadhyay

Ayan Mukherjee

Sundeep Khandelwal

Abhijan Bhattacharyya

Snehasis Banerjee

Anirban Dutta Choudhury

Sakyajit Bhattacharya

Vijayanand Mahadeo Banahatti

Prateep Misra

Debayan Mukherjee

Jaydip Sen

Shreyasi Datta

Himadri Sekhar Paul

Amit Swain

Utpal Garain

Dhaval Satish Jani

Tulika Bose

Joel Joseph

Abhisek Das

Gitesh Kulkarni

Jayeeta Mondal

Satanik Panda

Abhijan Bhattacharya

Trisrota Deb

Soumadeep Saha

Aakash Tyagi

Indrajit Bhattachyarya

This disclosure relates generally to method and system for an adaptive filter based learning model for time series sensor signal classification on edge devices. The adaptive filter based learning model for time series sensor signal classification enables automated-computationally lightweight learning (significant reduction in computational resources) and inferring/classification in real-time or near-real-time on CPU/memory/battery life constrained edge devices. The disclosed techniques for time series sensor signal classification on edge devices characterizes the intrinsic signal processing properties of the input time series sensor signals using linear adaptive filtering and derivative spectrum to efficiently construct the adaptive filter based learning model based on standard classification algorithms for time series sensor signal classification.

Adaptive filter based learning model for time series sensor signal classification on edge devices

Small and compact Deep Learning models are required for embedded AI in several domains. In many industrial use-cases, there are requirements to transform already trained models to ensemble embedded systems or re-train those for a given deployment scenario, with limited data for transfer learning. Moreover, the hardware platforms used in embedded application include FPGAs, AI hardware accelerators, System-on-Chips and on-premises computing elements (Fog/Network Edge). These are interconnected through heterogenous bus/network with different capacities. Method of the present disclosure finds how to automatically partition a given DNN into ensemble devices, considering the effect of accuracy—latency power—tradeoff, due to intermediate compression and effect of quantization due to conversion to AI accelerator SDKs. Method of the present disclosure is an iterative approach to obtain a set of partitions by repeatedly refining the partitions and generating a cascaded model for inference and training on ensemble hardware.

Combining compression, partitioning and quantization of DL models for fitment in hardware processors

This disclosure relates generally to a system and a method for mitigating generalization loss in deep neural network for time series classification. In an embodiment, the disclosed method includes compute an entropy of a timeseries training dataset, and a mean and a variance of the entropy and a regularization factor is computed. A plurality of iterations are performed to dynamically adjust the learning rate of the deep Neural Network (DNN) using a Mod-Adam optimization, and obtain a network parameter, and based on the network parameter, the regularization factor is updated to obtain an updated regularized factor. The learning rate is adjusted in the plurality of iterations by repeatedly updating the network parameter based on a variation of a generalization loss during the plurality of iterations. The updated regularized factor of the current iteration is used for adjusting the learning rate in a subsequent iteration of the plurality of iterations.

System and method for mitigating generalization loss in deep neural network for time series classification

This disclosure relates generally to multi-class multi-label classification and more particularly to contradiction avoided learning for multi-class multi-label classification. Conventional classification methods do not consider contradictory outcomes in multi-label classification tasks wherein contradictory outcomes have significant negative impact in the classification problem solution. The present disclosure provides a contradiction avoided learning multi-class multi-label classification. The disclosed method utilizes a binary contradiction matrix constructed using domain knowledge. Based on the binary contradiction matrix the training dataset is divided into two parts, one comprising contradictions and the second without contradictions. The classification model is trained using the divided datasets using a contradiction loss and a binary cross entropy loss to avoid contradictions during learning of the classification model. The disclosed method is used for electrocardiogram classification, shape classification and so on.

Method and system for contradiction avoided learning for multi-class multi-label classification

This disclosure relates generally to methods and systems for unobtrusive digital health assessment of high risk subjects, wherein bio-markers pertaining to a disease are identified automatically using physical activity and physiology monitoring on a continuous basis. Identification of bio-markers in the medical domain is conventionally dependent on insights derived from medical tests which are obtrusive in nature. Systems and methods of the present disclosure integrate physical characteristics, lifestyle habits and prevailing medical conditions with monitored physical activities and physiological measurements to assess health of high risk subjects. Systems and methods of the present disclosure also enable automatic generation of control class and treatment class that may be effectively used for health assessment.

Systems and methods for unobtrusive digital health assessment

This disclosure relates generally to physiological monitoring, and more particularly to feature set optimization for classification of physiological signal. In one embodiment, a method for physiological monitoring includes identifying clean physiological signal training set from an input physiological signal based on a Dynamic Time Warping (DTW) of segments associated with the physiological signal. An optimal features set is extracted from a clean physiological signal training set based on a Maximum Consistency and Maximum Dominance (MCMD) property associated with the optimal feature set that strictly optimizes on the objective function, the conditional likelihood maximization over different selection criteria such that diverse properties of different selection parameters are captured and achieves Pareto-optimality. The input physiological signal is classified into normal signal components and abnormal signal components using the optimal features set.

System and method for physiological monitoring and feature set optimization for classification of physiological signal

In many real-life applications, ample amount of examples from one class are present while examples from other classes are rare for training and learning purposes leading to class imbalance problem and misclassification. Methods and systems of the present disclosure facilitate generation of an extended synthetic rare class super dataset that is further pruned to obtain a synthetic rare class dataset by maximizing similarity and diversity in the synthetic rare class dataset while preserving morphological identity with labeled rare class training dataset. Oversampling methods used in the art result in cloning of datasets and do not provide the needed diversity. The methods of the present disclosure can be applied to classification of noisy phonocardiogram (PCG) signals among other applications.

Synthetic rare class generation by preserving morphological identity

A method and system for a feature subset-classifier pair for a classification task. The classification task corresponds to automatically classifying data associated with a subject(s) or object(s) of interest into an appropriate class based on a feature subset selected among a plurality of features extracted from the data and a classifier selected from a set of classifier types. The method proposed includes simultaneously determining the feature subset-classifier pair based on a relax-greedy {feature subset, classifier} approach utilizing sub-greedy search process based on a patience function, wherein the feature subset-classifier pair provides an optimal combination for more accurate classification. The automatic joint selection is time efficient solution, effectively speeding up the classification task.

Method and system for joint selection of a feature subset-classifier pair for a classification task

Systems and methods for detecting an anomaly in a cardiovascular signal using hierarchical extremas and repetitions. The traditional systems and methods provide for some anomaly detection in the cardiovascular signal but do not consider the discrete nature and strict rising and falling patterns of the cardiovascular signal and frequency in terms of hierarchical maxima points and minima points. Embodiments of the present disclosure provide for detecting the anomaly in the cardiovascular signal using hierarchical extremas and repetitions by smoothening the cardiovascular signal, deriving sets of hierarchical extremas using window detection, identifying signal patterns based upon the sets of hierarchical extremas, identifying repetitions in the signal patterns based upon occurrences and randomness of occurrences of the signal patterns and classifying the cardiovascular signal as anomalous and non-anomalous for detecting the anomaly in the cardiovascular signal.

Systems and methods for detecting anomaly in a cardiovascular signal using hierarchical extremas and repetitions

The present disclosure addresses the technical problem of information loss while representing a physiological signal in the form of symbols and for recognizing patterns inside the signal. Thus making it difficult to retain or extract any relevant information which can be used to detect anomalies in the signal. A system and method for anomaly detection and discovering pattern in a signal using morphology aware symbolic representation has been provided. The system discovers pattern atoms based on the strictly increasing and strictly decreasing characteristics of the time series physiological signal, and generate symbolic representation in terms of these pattern atoms. Additionally the method possess more generalization capability in terms of granularity. This detects discord/abnormal phenomena with consistency.

Method and system for pattern recognition in a signal using morphology aware symbolic representation

This disclosure relates generally to the use of distributed system for computation, and more particularly, relates to a method and system for optimizing computation and communication resource while preserving security in the distributed device for computation. In one embodiment, a system and method of utilizing plurality of constrained edge devices for distributed computation is disclosed. The system enables integration of the edge devices like residential gateways and smart phone into a grid of distributed computation. The edged devices with constrained bandwidth, energy, computation capabilities and combination thereof are optimized dynamically based on condition of communication network. The system further enables scheduling and segregation of data, to be analyzed, between the edge devices. The system may further be configured to preserve privacy associated with the data while sharing the data between the plurality of devices during computation.

System and method for distributed computation using heterogeneous computing nodes

Current technologies analyze electrocardiogram (ECG) signals for a long duration, which is not always a practical scenario. Moreover the current scenarios perform a binary classification between normal and Atrial Fibrillation (AF) only, whereas there are many abnormal rhythms apart from AF. Conventional systems/methods have their own limitations and may tend to misclassify ECG signals, thereby resulting in an unbalanced multi-label classification problem. Embodiments of the present disclosure provide systems and methods that are robust and more efficient for classifying rhythms for example, normal, AF, other abnormal rhythms and noisy ECG recordings by implementing a spectrogram based noise removal that obtains clean ECG signal from an acquired single-lead ECG signal, an optimum feature selection at each layer of classification that selects optimum features from a pool of extracted features, and a multi-layer cascaded binary classifier that identifies rhythms in the clean ECG signal at each layer of the classifier.

Cascaded binary classifier for identifying rhythms in a single-lead electrocardiogram (ECG) signal

Accurate detection of anomaly in sensor signals is critical and can have an immense impact in the health care domain. Accordingly, identifying outliers or anomalies with reduced error and reduced resource usage is a challenge addressed by the present disclosure. Self-learning of normal signature of an input sensor signal is used to derive primary features based on valley and peak points of the sensor signals. A pattern is recognized by using discrete nature and strictly rising and falling edges of the input sensor signal. One or more defining features are identified from the derived features based on statistical properties and time and frequency domain properties of the input sensor signal. Based on the values of the defining features, clusters of varying density are identified for the input sensor signal and based on the density of the clusters, anomalous and non-anomalous portions of the input sensor signals are classified.

Anomaly detection by self-learning of sensor signals

The present subject matter discloses a system and a method for identifying information from sensor data in a sensor agnostic manner. The system may receive sensor data provided by a sensor and may determine statistical features of the sensor data. The system may determine signal dynamics of the sensor data based on at least one of the statistical features, signal processing features, and a data distribution model. The system may select at least one outlier class based on the signal dynamics, number of streams of the sensor data, and dimensions of the sensor data. The system may select at least one outlier detection method associated with an outlier class for detecting outliers in the sensor data. The system may determine information content of the sensor data based on the outliers, the signal dynamics, the statistical features, and information theoretic features, and similarity or dissimilarity measure.

System and method for determining information and outliers from sensor data

Disclosed is a method and system for detecting outliers in real-time for a univariate time-series signal. The system may receive the univariate time-series signal, comprising a plurality of datasets, from a data source. The system may compute a standard deviation of a dataset of the plurality of datasets. Subsequently, the system may compute the optimal sample block size and the critical sample size of the dataset. Further, the system may determine the optimal operational block size of the dataset. The system may segment the plurality of datasets into blocks based upon the optimal operational block size. The system may detect the outliers by performing an outlier detection technique on the blocks, thereby ensuring improved execution time while minimally affecting precision and accuracy of the outcome of the outlier detection method.

Growing community of inventors

Kolkata, India

Arijit Ukil