Tata Consultancy Services Limited

Gautam Shroff

Lovekesh Vig

Puneet Kumar Agarwal

Pankaj Malhotra

Ashwin Srinivasan

Mayur Patidar

Priyanka Gupta

Garima Gupta

Puneet Agarwal

Karamjit Singh

Geetika Sharma

Aditeya Pandey

Prerna Khurana

Gunjan Sehgal

Mrinal Rawat

Arpan Pal

Rekha Singhal

Lipika Dey

Hiranmay Ghosh

Amit Sangroya

Narendhar Gugulothu

Diksha Garg

Jyoti Narwariya

Sarmimala Saikia

Kaushal Ashokbhai Paneri

Vishal Sunder

Aniruddha Sinha

Venkataramana Runkana

Monika Sharma

Ranjan Dasgupta

Shruti Kunde

Kingshuk Chakravarty

Manoj Karunakara Nambiar

Arindam Saha

Aditya Pareek

Vishnu Swaroopji Masampally

Ehtesham Hassan

Sivakumar Subramanian

Dheeraj Chahal

Manasi Samarth Patwardhan

Rajat Kumar

Shubham Paliwal

Rohit Rahul

Arindam Chowdhury

Ishan Verma

Mayank Mishra

Ankit Sharma

Pratik Saini

Chandrasekhar Anantaram

Rupinder Singh Virk

Rohit Kumar

Sharod Roy Choudhury

Vivek Bisht

Arnab Chatterjee

Vishnu Tv

Prateek Sachan

Mouli Rastogi

Vaibhav Varshney

Manasi Malik

Vishnu Tankasala Veparala

Ankit Bansal

Nishit Raj

Siddharth Verma

Ashish Chaudhary

Rachit Bansal

Vishnu T V

Shabbirhussain Hamid Bhaisaheb

Lovkesh Vig

Solomon Pushparaj Manuelraj

Syed Afshan Ali

Ranjitha Prasad

Mohit Yadav

Rajaswa Ravindra Patil

Rishabh Gupta

Brijendra Singh

Rajgopal Vaithiyanathan

Somya Singh

Vibhor Gupta

Arpit Khurdiya

Vishwanath Doreswamy

Diwakar Mahajan

Tavleen Oberoi

Sujal Subhash Wattamwar

Kingshuk Chakravarhy

Vartika Tewari

Mahesh Prasad Singh

Sanket Kadarkar

Surya Yadav

Richa Rawat

This disclosure relates generally to program synthesis for weakly-supervised multimodal question answering using filtered iterative back-translation (FIBT). Existing approaches for chart question answering mainly address structural, visual, relational, or simple data retrieval queries with fixed-vocabulary answers. The present disclosure implements a two-stage approach where, in first stage, a computer vision pipeline is employed to extract data from chart images and store in a generic schema. In second stage, SQL programs for Natural Language (NL) queries are generated in dataset by using FIBT. To adapt forward and backward models to required NL queries, a Probabilistic Context-Free Grammar is defined, whose probabilities are set to be inversely proportional to SQL programs in training data and sample programs from it. Compositional similarity-based filtration strategy employed on the NL queries generated for these SQL programs enables synthesizing, filtering, and appending NL query-SQL program pairs to training data, iteratively moving towards required NL query distribution.

System and method for program synthesis for weakly-supervised multimodal question answering using filtered iterative back-translation

Existing semi-supervised and unsupervised approaches for intent discovery require an estimate of the number of new intents present in the user logs. The present disclosure receives labeled utterances from known intents and update parameters of a pre-trained language model (PLM). Representation learning and clustering is performed iteratively using labeled and unlabeled utterances from known intents and unlabeled utterances from unknown intents to fine-tune PLM and a plurality of clusters is generated. Cluster merger algorithm is executed iteratively on generated plurality of clusters. A query cluster is obtained by randomly selecting one cluster from the plurality of clusters and by obtaining a corresponding plurality of nearest neighbors based on a cosine-similarity. A response for merging the query cluster and corresponding plurality of nearest neighbors is obtained, and a new cluster is created. The corresponding cluster representation is recalculated and each of the new cluster is interpreted as an intent.

System and method for intent discovery from user logs using deep semi-supervised contrastive clustering

Recommender Systems (RS) tend to recommend more popular items instead of the relevant long-tail items. Mitigating such popularity bias is crucial to ensure that less popular but relevant items are recommended. System described herein analyses popularity bias in session-based RS obtained via deep learning (DL) models. DL models trained on historical user-item interactions in session logs (having long-tailed item-click distributions) tend to amplify popularity bias. To understand source of this bias amplification, potential sources of bias at data-generation stage (user-item interactions captured as session logs) and model training stage are considered by the system for recommendation wherein popularity of item has causal effect on user-item interactions via conformity bias, and item ranking from models via biased training process due to class imbalance. While most existing approaches address only one of these effects, a comprehensive causal inference framework is implemented by present disclosure that identifies and mitigates effects at both stages.

System, method, and non-transitory machine-readable information storage medium for recommendations of items and controlling an associated bias thereof

Session-based Recommendation (SR) is the task of recommending the next item based on previously recorded user interactions. However, most existing approaches for SR either rely on costly online interactions with real users (model-free approaches) or rely on potentially biased rule-based or data-driven user-behavior models (model-based approaches) for learning. This disclosure relates to a system and method for selecting session-based recommendation policies using historical recommendations and user feedback. Herein, the learning of recommendation policies given offline or batch data from old recommendation policies based on a Distributional Reinforcement Learning (DRL) based recommender system in the offline or batch-constrained setting without requiring access to a user-behavior model or real-interactions with the users.

System and method for training recommendation policies

Several applications capture data from sensors resulting in multi-sensor time series. Existing neural networks-based approaches for such multi-sensor/multivariate time series modeling assume fixed input-dimension/number of sensors. Such approaches can struggle in practical setting where different instances of same device/equipment come with different combinations of installed sensors. In the present disclosure, neural network models are trained from such multi-sensor time series having varying input dimensionality, owing to availability/installation of different sensors subset at each source of time series. Neural network (NN) architecture is provided for zero-shot transfer learning allowing robust inference for multivariate time series with previously unseen combination of available dimensions/sensors at test time. Such combinatorial generalization is achieved by conditioning layers of core NN-based time series model with 'conditioning vector' carrying information of available sensors combination for each time series and is obtained by summarizing learned “sensor embedding vectors set” corresponding to available sensors in time series.

Neural networks for handling variable-dimensional time series data

Existing techniques assume that all time varying covariates are confounding and thus attempts to balance a full state representation of a plurality of historical observants. The present disclosure processes a plurality of historical observants and treatment at a timestep t specific to each patient using an encoder network to a obtain a state representation s. A first set of disentangled representations comprising an outcome, a confounding and a treatment representation is learnt to predict an outcome. The first set of disentangled representations are concatenated to obtain a unified representation and the decoder network is initialized using the unified representation to obtain a state representation s. A second set of disentangled representations is learnt and concatenated to predict outcomem+1 timesteps ahead of the timestep t and proceeding iteratively until m=τ−1.

System and method for learning disentangled representations for temporal casual inference

This disclosure relates generally to method and system for handling popularity bias in item recommendations. In an embodiment the method includes initializing an item embedding look-up matrix corresponding to items in a sequence of item-clicks with respect to a training data. L2 norm is applied to the item embedding look-up matrix to learn a normalized item embeddings. Using a neural network, a session embeddings corresponding to the sequences of item-clicks is modeled and L2 norm is applied to the session embeddings to obtain a normalized session embeddings. Relevance scores corresponding to each of the plurality of items are obtained based on similarity between the normalized item embeddings and the normalized session embeddings. A multi-dimensional probability vector corresponding to the relevance scores for the items to be clicked in the sequence is obtained. A list of the items ordered based on the multi-dimensional probability vector is provided as recommendation.

System, method, and non-transitory machine readable information storage medium for handling popularity bias in item recommendations

This disclosure relates to optimizing an operation of an equipment by a neural network based optimizer is provided. The method include receiving, information associated with at least one equipment instance (j) as an input at a predefined sequence of timestamps; training, a plurality of simulation models for each equipment instance to obtain a function (f); processing, the external input parameters (e) to obtain a fixed-dimensional vector and passed as an input to obtain an vector (i); generating, a modified (i) from the output vector (i) based on a domain constraint value; computing, a reward (r) based on (i) the function (f), (ii) the modified (i), (iii) the external input parameters (e), and (iv) a reward function (R); and iteratively performing the steps of processing, generating, and computing reward (r) for a series of subsequent equipment instances after expiry of the predefined sequence of timestamps associated with a first equipment instance.

System and method for operation optimization of an equipment

In the world of digital advertising, optimally allocating an advertisement campaign within a fixed pre-defined budget for an advertising duration aimed at maximizing number of conversions is very important for an advertiser. Embodiments of present disclosure provides a robust and easily generalizable method of optimal allocation of advertisement campaign by formulating it as a constrained Markov Decision Process (MDP) defined by agent state comprising user state and advertiser state, action space comprising a plurality of ad campaigns, state transition routine and a cumulative reward model which rewards maximum total conversions in an advertising duration. The cumulative reward model is trained in conjunction with a deep Q-network for solving the MDP to optimally allocate advertisement campaign for an advertising duration within a constrained budget.

Budget constrained deep Q-network for dynamic campaign allocation in computational advertising

Organizations are constantly flooded with questions, ranging from mundane to the unanswerable. It is therefore respective department that actively looks for automated assistance, especially to alleviate the burden of routine, but time-consuming tasks. The embodiments of the present disclosure provide BiLSTM-Siamese Network based Classifier for identifying target class of queries and providing responses to queries pertaining to the identified target class, which acts as an automated assistant that alleviates burden of answering queries in well-defined domains. Siamese Model (SM) is trained for a epochs, and then the same Base-Network is used to train Classification Model (CM) for b epochs iteratively until best accuracy is observed on validation test, wherein SM ensures it learns which sentences are similar/dissimilar semantically while CM learns to predict target class of every user query. Here a and b are assumed to be hyper parameters and are tuned for best performance on the validation set.

BiLSTM-siamese network based classifier for identifying target class of queries and providing responses thereof

This disclosure relates to a method and system for extracting information from images of one or more templatized documents. A knowledge graph with a fixed schema based on background knowledge is used to capture spatial and semantic relationships of entities present in scanned document and an adaptive lattice-based approach based on formal concepts analysis (FCA) is used to determine a similarity metric that utilizes both spatial and semantic information to determine if the structure of the scanned document image adheres to any of the known document templates. If a known document template whose structure is closely matching the structure of the scanned document is detected, then an inductive rule learning based approach is used to learn symbolic rules to extract information present in scanned document image and if a new document template is detected, then future scanned document images belonging to new document template are automatically processed using the learnt rules.

Methods and systems for extracting information from document images

For various applications (for example, a Virtual Assistant), mechanisms that are capable of collecting user queries and generating responses are being used. While such systems handle structured queries well, they struggle to or fail to interpret an unstructured Natural Language (NL) query. The disclosure herein generally relates to data processing, and, more particularly, to a method and a system for generating responses to unstructured Natural Language (NL) queries. The system collects at least one NL query as input at a time, and generates a sketch, where the sketch is a structured representation of the unstructured NL query. Further by processing the sketch, the system generates one or more database queries. The one or more database queries are then used to search in one or more associated databases and to retrieve matching results, which are then used to generate response to the at least one NL query.

Method and system for generating a response to an unstructured natural language (NL) query

Neural networks can be used for time series data classification. However, in a K-shot scenario in which sufficient training data is unavailable to train the neural network, the neural network may not produce desired results. Disclosed herein are a method and system for training a neural network for time series data classification. In this method, by processing a plurality of task specific data, a system generates a set of updated parameters, which is further used to train a neural network (network) till a triplet loss is below a threshold. The network is trained on a diverse set of few-shot tasks sampled from various domains (e.g. healthcare, activity recognition, and so on) such that it can solve a target task from another domain using only a small number of training samples from the target task.

Method and system for training a neural network for time series data classification

Estimating Remaining Useful Life (RUL) from multi-sensor time series data is difficult through manual inspection. Current machine learning and data analytics methods, for RUL estimation require large number of failed instances for training, which are rarely available in practice, and these methods cannot use information from currently operational censored instances since their failure time is unknown. Embodiments of the present disclosure provide systems and methods for estimating RUL using time series data by implementing an LSTM-RNN based ordinal regression technique, wherein during training RUL value of failed instance(s) is encoded into a vector which is given as a target to the model. Unlike a failed instance, the exact RUL for a censored instance is unknown. For using the censored instances, target vectors are generated and the objective function is modified for training wherein the trained LSTM-RNN based ordinal regression is applied on an input test time series for RUL estimation.

Failed and censored instances based remaining useful life (RUL) estimation of entities

Users have to assign labels to a ticket to route to right domain expert for resolving issue(s). In practice, labels are large and organized in form of a tree. Lack in clarity in problem description has resulted in inconsistent and incorrect labeling of data, making it hard for one to learn/interpret. Embodiments of the present disclosure provide systems and methods that identify relevant queries to obtain user response, for identification of right category and ticket logging there. This is achieved by implementing attention based sequence to sequence (seq2seq) hierarchical classification model to assign the hierarchical categories to tickets, followed by a slot filling model to enable identifying/deciding right set of queries, if the top-k model predictions are not consistent. Further, training data for slot filling model is automatically generated based on attention weight in the hierarchical classification model.

Systems and methods for conversational based ticket logging

Growing community of inventors

Gurgaon, India

Gautam Shroff