no code implementations • NAACL 2022 • Rajat Kumar, Mayur Patidar, Vaibhav Varshney, Lovekesh Vig, Gautam Shroff
However, even skilled domain experts are often unable to foresee all possible user intents at design time and for practical applications, novel intents may have to be inferred incrementally on-the-fly from user utterances.
Ranked #1 on Open Intent Discovery on BANKING77
no code implementations • Findings (NAACL) 2022 • Vaibhav Varshney, Mayur Patidar, Rajat Kumar, Lovekesh Vig, Gautam Shroff
This typically entails repeated retraining of the intent detector on both the existing and novel intents which can be expensive and would require storage of all past data corresponding to prior intents.
no code implementations • EACL (AdaptNLP) 2021 • Surabhi Kumari, Nikhil Jaiswal, Mayur Patidar, Manasi Patwardhan, Shirish Karande, Puneet Agarwal, Lovekesh Vig
In comparison, in this work, we observe that a simpler filtering approach based on a domain classifier, applied only to the pseudo-training data can consistently perform better, providing performance gains of 1. 40, 1. 82 and 0. 76 in terms of BLEU score for Medical, Law and IT in one direction, and 1. 28, 1. 60 and 1. 60 in the other direction in low resource scenario over competitive baselines.
no code implementations • AACL (WAT) 2020 • Nikhil Jaiswal, Mayur Patidar, Surabhi Kumari, Manasi Patwardhan, Shirish Karande, Puneet Agarwal, Lovekesh Vig
This is further followed by fine-tuning on the domain-specific corpus.
no code implementations • 7 Mar 2024 • Harshit Nigam, Manasi Patwardhan, Lovekesh Vig, Gautam Shroff
To aid with research ideation, we propose `Acceleron', a research accelerator for different phases of the research life cycle, and which is specially designed to aid the ideation process.
no code implementations • 23 Nov 2023 • Ritam Majumdar, Amey Varhade, Shirish Karande, Lovekesh Vig
Physics Informed Neural Operators (PINO) overcome this constraint by utilizing a physics loss for the training, however the accuracy of PINO trained without data does not match the performance obtained by training with data.
no code implementations • 18 Aug 2023 • Ritam Majumdar, Vishal Jadhav, Anirudh Deodhar, Shirish Karande, Lovekesh Vig, Venkataramana Runkana
Physics-informed neural networks (PINNs) have been widely used to develop neural surrogates for solutions of Partial Differential Equations.
no code implementations • 18 Aug 2023 • Ritam Majumdar, Shirish Karande, Lovekesh Vig
Simulating physical systems using Partial Differential Equations (PDEs) has become an indispensible part of modern industrial process optimization.
no code implementations • 1 Aug 2023 • Aseem Arora, Shabbirhussain Bhaisaheb, Harshit Nigam, Manasi Patwardhan, Lovekesh Vig, Gautam Shroff
Cross-domain and cross-compositional generalization of Text-to-SQL semantic parsing is a challenging task.
no code implementations • 26 Apr 2023 • Krishnam Hasija, Shrishti Pradhan, Manasi Patwardhan, Raveendra Kumar Medicherla, Lovekesh Vig, Ravindra Naik
We further fine-tune UnixCoder, the best-performing model for zero-shot cross-programming language code search, for the Code Cloning task with the SBT IRs of C code-pairs, available in the CodeNet dataset.
no code implementations • 24 Mar 2023 • Ritam Majumdar, Shirish Karande, Lovekesh Vig
We then use a neural network to learn the mapping between spread trajectories and coefficients of SIDR in an offline manner.
no code implementations • 16 Mar 2023 • Ankita Sontakke, Kanika Kalra, Manasi Patwardhan, Lovekesh Vig, Raveendra Kumar Medicherla, Ravindra Naik, Shrishti Pradhan
In this paper, we focus on transferring the knowledge acquired by the code-to-pseudocode neural model trained on a high resource PL (C++) using parallel code-pseudocode data.
no code implementations • 13 Mar 2023 • Ritam Majumdar, Vishal Jadhav, Anirudh Deodhar, Shirish Karande, Lovekesh Vig, Venkataramana Runkana
Furthermore, on an average, pruning improves the accuracy of DPA by 7. 81% .
no code implementations • 4 Mar 2023 • Shreyas Bhat Brahmavar, Rohit Rajesh, Tirtharaj Dash, Lovekesh Vig, Tanmay Tulsidas Verlekar, Md Mahmudul Hasan, Tariq Khan, Erik Meijering, Ashwin Srinivasan
Deep neural network (DNN) models for retinopathy have estimated predictive accuracies in the mid-to-high 90%.
1 code implementation • 20 Feb 2023 • Soham Rohit Chitnis, Sidong Liu, Tirtharaj Dash, Tanmay Tulsidas Verlekar, Antonio Di Ieva, Shlomo Berkovsky, Lovekesh Vig, Ashwin Srinivasan
To investigate the effect of domain-specific pre-training, we considered the current state-of-the-art multiple-instance learning models, 1) CLAM, an attention-based model, and 2) TransMIL, a self-attention-based model, and evaluated the models' confidence and predictive performance in detecting primary brain tumors - gliomas.
no code implementations • 15 Jan 2023 • S I Harini, Gautam Shroff, Ashwin Srinivasan, Prayushi Faldu, Lovekesh Vig
We model short-duration (e. g. day) trading in financial markets as a sequential decision-making problem under uncertainty, with the added complication of continual concept-drift.
no code implementations • 20 Dec 2022 • Ramya Hebbalaguppe, Rishabh Patra, Tirtharaj Dash, Gautam Shroff, Lovekesh Vig
Contemporary model calibration techniques mitigate the problem of overconfident predictions by pushing down the confidence of the winning class while increasing the confidence of the remaining classes across all test samples.
no code implementations • 20 Dec 2022 • Ritam Majumdar, Vishal Jadhav, Anirudh Deodhar, Shirish Karande, Lovekesh Vig, Venkataramana Runkana
We demonstrate a Physics-informed Neural Network (PINN) based model for real-time health monitoring of a heat exchanger, that plays a critical role in improving energy efficiency of thermal power plants.
no code implementations • 20 Dec 2022 • Mayur Patidar, Prayushi Faldu, Avinash Singh, Lovekesh Vig, Indrajit Bhattacharya, Mausam
When answering natural language questions over knowledge bases, missing facts, incomplete schema and limited scope naturally lead to many questions being unanswerable.
no code implementations • 29 Nov 2022 • Vedant Shah, Aditya Agrawal, Lovekesh Vig, Ashwin Srinivasan, Gautam Shroff, Tanmay Verlekar
Real data is often imperfect meaning that even if the symbolic theory remains unchanged, we may still need to address the problem of mapping raw data to high-level symbols, each time there is a change in the data acquisition environment or equipment.
no code implementations • 19 Sep 2022 • Vishwa Shah, Aditya Sharma, Gautam Shroff, Lovekesh Vig, Tirtharaj Dash, Ashwin Srinivasan
However, connectionist models struggle to include explicit domain knowledge for deductive reasoning.
no code implementations • 11 Jul 2022 • Ritam Majumdar, Vishal Jadhav, Anirudh Deodhar, Shirish Karande, Lovekesh Vig, Venkataramana Runkana
We use this approximation to define multilayer symbolic networks.
no code implementations • 30 Mar 2022 • Manu Sheoran, Meghal Dani, Monika Sharma, Lovekesh Vig
Existing universal lesion detection (ULD) methods utilize compute-intensive anchor-based architectures which rely on predefined anchor boxes, resulting in unsatisfactory detection performance, especially in small and mid-sized lesions.
Ranked #6 on Medical Object Detection on DeepLesion
no code implementations • 14 Mar 2022 • Arushi Jain, Shubham Paliwal, Monika Sharma, Lovekesh Vig
In this paper, we propose to train a deep network to capture the spatial associations between different word pairs present in the table image for unravelling the table structure.
no code implementations • British Machine Vision Conference 2021 • Manu Sheoran, Meghal Dani, Monika Sharma, Lovekesh Vig
In this paper, we exploit the domain information present in computed tomography (CT) scans and propose a robust universal lesion detection (ULD) network that can detect lesions across all organs of the body by training on a single dataset, DeepLesion.
Ranked #2 on Medical Object Detection on DeepLesion
no code implementations • 14 Mar 2022 • Vibhor Gupta, Jyoti Narwariya, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
We note that existing continual learning methods do not take into account variability in input dimensions arising due to different subsets of sensors being available across tasks, and struggle to adapt to such variable input dimensions (VID) tasks.
no code implementations • 25 Feb 2022 • Diksha Garg, Pankaj Malhotra, Anil Bhatia, Sanjay Bhat, Lovekesh Vig, Gautam Shroff
We consider learning a trading agent acting on behalf of the treasury of a firm earning revenue in a foreign currency (FC) and incurring expenses in the home currency (HC).
no code implementations • 11 Feb 2022 • Jyoti Narwariya, Chetan Verma, Pankaj Malhotra, Lovekesh Vig, Easwara Subramanian, Sanjay Bhat
One of the objectives of such demand response management is to incentivize the consumers to adjust their consumption so that the overall electricity procurement in the wholesale markets is minimized, e. g. it is desirable that consumers consume less during peak hours when cost of procurement for brokers from wholesale markets are high.
no code implementations • 20 Jan 2022 • Garima Gupta, Lovekesh Vig, Gautam Shroff
Medical professionals evaluating alternative treatment plans for a patient often encounter time varying confounders, or covariates that affect both the future treatment assignment and the patient outcome.
no code implementations • 19 Nov 2021 • Atharv Sonwane, Gautam Shroff, Lovekesh Vig, Ashwin Srinivasan, Tirtharaj Dash
We consider a class of visual analogical reasoning problems that involve discovering the sequence of transformations by which pairs of input/output images are related, so as to analogously transform future inputs.
no code implementations • 31 Oct 2021 • Mrinal Rawat, Ramya Hebbalaguppe, Lovekesh Vig
While Out-of-distribution (OOD) detection has been well explored in computer vision, there have been relatively few prior attempts in OOD detection for NLP classification.
no code implementations • 19 Oct 2021 • Atharv Sonwane, Sharad Chitlangia, Tirtharaj Dash, Lovekesh Vig, Gautam Shroff, Ashwin Srinivasan
The ability to solve Bongard problems is an example of such a test.
no code implementations • 8 Sep 2021 • Shubham Paliwal, Monika Sharma, Lovekesh Vig
The proposed pipeline, named OSSR-PID, is fast and gives outstanding performance for recognition of symbols on a synthetic dataset of 100 P&ID diagrams.
no code implementations • 8 Sep 2021 • Shubham Paliwal, Arushi Jain, Monika Sharma, Lovekesh Vig
A novel and efficient kernel-based line detection and a two-step method for detection of complex symbols based on a fine-grained deep recognition technique is presented in the paper.
no code implementations • 21 Dec 2020 • Sachin Kumar, Garima Gupta, Ranjitha Prasad, Arnab Chatterjee, Lovekesh Vig, Gautam Shroff
Advertising channels have evolved from conventional print media, billboards and radio advertising to online digital advertising (ad), where the users are exposed to a sequence of ad campaigns via social networks, display ads, search etc.
no code implementations • 19 Dec 2020 • Rishab Khincha, Soundarya Krishnan, Tirtharaj Dash, Lovekesh Vig, Ashwin Srinivasan
In this paper, deep neural networks are used to extract domain-specific features(morphological features like ground-glass opacity and disease indications like pneumonia) directly from the image data.
no code implementations • 16 Dec 2020 • Diksha Garg, Priyanka Gupta, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
Most of the existing deep reinforcement learning (RL) approaches for session-based recommendations either rely on costly online interactions with real users, or rely on potentially biased rule-based or data-driven user-behavior models for learning.
2 code implementations • 23 Oct 2020 • Tirtharaj Dash, Ashwin Srinivasan, Lovekesh Vig
These kinds of problems have been addressed effectively in the past by Inductive Logic Programming (ILP), by virtue of 2 important characteristics: (a) The use of a representation language that easily captures the relation encoded in graph-structured data, and (b) The inclusion of prior information encoded as domain-specific relations, that can alleviate problems of data scarcity, and construct new relations.
no code implementations • 22 Aug 2020 • Ankit Sharma, Garima Gupta, Ranjitha Prasad, Arnab Chatterjee, Lovekesh Vig, Gautam Shroff
The proposed architecture comprises of a decorrelation network and an outcome prediction network.
no code implementations • 1 Jul 2020 • Vibhor Gupta, Jyoti Narwariya, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
Such a combinatorial generalization is achieved by conditioning the layers of a core neural network-based time series model with a "conditioning vector" that carries information of the available combination of sensors for each time series.
no code implementations • 30 Jun 2020 • Jyoti Narwariya, Pankaj Malhotra, Vishnu Tv, Lovekesh Vig, Gautam Shroff
Deep learning models such as those based on recurrent neural networks (RNNs) or convolutional neural networks (CNNs) fail to explicitly leverage this potentially rich source of domain-knowledge into the learning procedure.
no code implementations • 28 Apr 2020 • Ankit Sharma, Garima Gupta, Ranjitha Prasad, Arnab Chatterjee, Lovekesh Vig, Gautam Shroff
Causal inference (CI) in observational studies has received a lot of attention in healthcare, education, ad attribution, policy evaluation, etc.
5 code implementations • 6 Jan 2020 • Shubham Paliwal, Vishwanath D, Rohit Rahul, Monika Sharma, Lovekesh Vig
This includes accurate detection of the tabular region within an image, and subsequently detecting and extracting information from the rows and columns of the detected table.
Ranked #3 on Table Detection on ICDAR2013
no code implementations • 9 Dec 2019 • Ankit Sharma, Garima Gupta, Ranjitha Prasad, Arnab Chatterjee, Lovekesh Vig, Gautam Shroff
Performing inference on data obtained through observational studies is becoming extremely relevant due to the widespread availability of data in fields such as healthcare, education, retail, etc.
no code implementations • 21 Nov 2019 • Monika Sharma, Shikha Gupta, Arindam Chowdhury, Lovekesh Vig
To this end, we formulate the problem of reasoning over statistical charts as a classification task using MAC-Networks to give answers from a predefined vocabulary of generic answers.
no code implementations • 14 Nov 2019 • Kushagra Mahajan, Monika Sharma, Lovekesh Vig
The algorithm is both fast and accurate and utilizes a standard Optical character recognition (OCR) engine such as Tesseract to find character based unambiguous keypoints, which are utilized to identify precise keypoint correspondences between two images.
no code implementations • WS 2019 • Mayur Patidar, Surabhi Kumari, Manasi Patwardhan, Kar, Shirish e, Puneet Agarwal, Lovekesh Vig, Gautam Shroff
We observe that the proposed approach provides consistent gains in the performance of BERT for multiple benchmark datasets (e. g. 1. 0{\%} gain on MLDocs, and 1. 2{\%} gain on XNLI over translate-train with BERT), while requiring a single model for multiple languages.
no code implementations • 13 Sep 2019 • Jyoti Narwariya, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff, Vishnu Tv
We overcome this limitation in order to train a common agent across domains with each domain having different number of target classes, we utilize a triplet-loss based learning procedure that does not require any constraints to be enforced on the number of classes for the few-shot TSC tasks.
2 code implementations • 10 Sep 2019 • Priyanka Gupta, Diksha Garg, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
The models using normalized item and session-graph representations perform significantly better: i. for the less popular long-tail items in the offline setting, and ii.
Ranked #2 on Session-Based Recommendations on Last.FM
no code implementations • 16 Jul 2019 • Vishnu TV, Pankaj Malhotra, Jyoti Narwariya, Lovekesh Vig, Gautam Shroff
Recently, neural networks trained as optimizers under the "learning to learn" or meta-learning framework have been shown to be effective for a broad range of optimization tasks including derivative-free black-box function optimization.
no code implementations • 6 Jun 2019 • Vishal Sunder, Ashwin Srinivasan, Lovekesh Vig, Gautam Shroff, Rohit Rahul
Our interest in this paper is in meeting a rapidly growing industrial demand for information extraction from images of documents such as invoices, bills, receipts etc.
no code implementations • 29 Apr 2019 • Kathan Kashiparekh, Jyoti Narwariya, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
We also provide qualitative insights into the working of CTN by: i) analyzing the activations and filters of first convolution layer suggesting the filters in CTN are generically useful, ii) analyzing the impact of the design decision to incorporate multiple length decisions, and iii) finding regions of time series that affect the final classification decision via occlusion sensitivity analysis.
no code implementations • 1 Apr 2019 • Priyanka Gupta, Pankaj Malhotra, Jyoti Narwariya, Lovekesh Vig, Gautam Shroff
We, therefore, conclude that pre-trained deep models like TimeNet and HealthNet allow leveraging the advantages of deep learning for clinical time series analysis tasks, while also minimize dependence on hand-crafted features, deal robustly with scarce labeled training data scenarios without overfitting, as well as reduce dependence on expertise and resources required to train deep networks from scratch.
no code implementations • 23 Mar 2019 • Vishnu TV, Diksha, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
We propose a novel deep learning based approach for Prognostics with Uncertainty Quantification that is useful in scenarios where: (i) access to labeled failure data is scarce due to rarity of failures (ii) future operational conditions are unobserved and (iii) inherent noise is present in the sensor readings.
no code implementations • 28 Jan 2019 • Rohit Rahul, Shubham Paliwal, Monika Sharma, Lovekesh Vig
To that end, we present a novel pipeline for information extraction from P&ID sheets via a combination of traditional vision techniques and state-of-the-art deep learning models to identify and isolate pipeline codes, pipelines, inlets and outlets, and for detecting symbols.
no code implementations • 28 Jan 2019 • Monika Sharma, Abhishek Verma, Lovekesh Vig
We compare the performance of CycleGAN for document cleaning tasks using unpaired images with a Conditional GAN trained on paired data from the same dataset.
no code implementations • 28 Dec 2018 • Vishwanath D, Lovekesh Vig, Gautam Shroff, Puneet Agarwal
In this paper we present Meeting Bot, a reinforcement learning based conversational system that interacts with multiple users to schedule meetings.
no code implementations • 11 Dec 2018 • Vishwanath D, Rohit Rahul, Gunjan Sehgal, Swati, Arindam Chowdhury, Monika Sharma, Lovekesh Vig, Gautam Shroff, Ashwin Srinivasan
In this paper, we propose a novel enterprise based end-to-end framework called DeepReader which facilitates information extraction from document images via identification of visual entities and populating a meta relational model across different entities in the document image.
Optical Character Recognition Optical Character Recognition (OCR) +2
no code implementations • 19 Sep 2018 • Vishal Sunder, Lovekesh Vig, Arnab Chatterjee, Gautam Shroff
We further train a meta agent with a mixture of behaviors by learning an ensemble of different models using reinforcement learning.
Multi-agent Reinforcement Learning reinforcement-learning +1
no code implementations • 20 Jul 2018 • Arindam Chowdhury, Lovekesh Vig
Offline handwritten text recognition from images is an important problem for enterprises attempting to digitize large volumes of handmarked scanned documents/reports.
no code implementations • 4 Jul 2018 • Priyanka Gupta, Pankaj Malhotra, Lovekesh Vig, Gautam Shroff
We consider transferring the knowledge captured in an RNN trained on several source tasks simultaneously using a large labeled dataset to build the model for a target task with limited labeled data.
no code implementations • 2 Jul 2018 • Ashwin Srinivasan, Lovekesh Vig, Michael Bain
We investigate the use of a Bayes-like approach to identify logical proxies for local predictions of a DRM.
no code implementations • IJCNLP 2017 • S Vishal, Mohit Yadav, Lovekesh Vig, Gautam Shroff
We present a novel technique for segmenting chat conversations using the information bottleneck method (Tishby et al., 2000), augmented with sequential continuity constraints.
no code implementations • 4 Sep 2017 • Narendhar Gugulothu, Vishnu Tv, Pankaj Malhotra, Lovekesh Vig, Puneet Agarwal, Gautam Shroff
We propose Embed-RUL: a novel approach for RUL estimation from sensor data that does not rely on any degradation-trend assumptions, is robust to noise, and handles missing values.
2 code implementations • 23 Jun 2017 • Pankaj Malhotra, Vishnu Tv, Lovekesh Vig, Puneet Agarwal, Gautam Shroff
Inspired by the tremendous success of deep Convolutional Neural Networks as generic feature extractors for images, we propose TimeNet: a deep recurrent neural network (RNN) trained on diverse time series in an unsupervised manner using sequence to sequence (seq2seq) models to extract features from time series.
no code implementations • EACL 2017 • Mohit Yadav, Lovekesh Vig, Gautam Shroff
Motivated by these practical issues, we propose a novel curriculum inspired training procedure for Memory Networks to improve the performance for machine comprehension with relatively small volumes of training data.
no code implementations • 3 Jan 2017 • Karamjit Singh, Garima Gupta, Lovekesh Vig, Gautam Shroff, Puneet Agarwal
Discovering causal models from observational and interventional data is an important first step preceding what-if analysis or counterfactual reasoning.
no code implementations • 20 Dec 2016 • Sarmimala Saikia, Lovekesh Vig, Ashwin Srinivasan, Gautam Shroff, Puneet Agarwal, Richa Rawat
We investigate solving discrete optimisation problems using the estimation of distribution (EDA) approach via a novel combination of deep belief networks(DBN) and inductive logic programming (ILP). While DBNs are used to learn the structure of successively better feasible solutions, ILP enables the incorporation of domain-based background knowledge related to the goodness of solutions. Recent work showed that ILP could be an effective way to use domain knowledge in an EDA scenario. However, in a purely ILP-based EDA, sampling successive populations is either inefficient or not straightforward. In our Neuro-symbolic EDA, an ILP engine is used to construct a model for good solutions using domain-based background knowledge. These rules are introduced as Boolean features in the last hidden layer of DBNs used for EDA-based optimization. This incorporation of logical ILP features requires some changes while training and sampling from DBNs: (a)our DBNs need to be trained with data for units at the input layer as well as some units in an otherwise hidden layer, and (b)we would like the samples generated to be drawn from instances entailed by the logical model. We demonstrate the viability of our approach on instances of two optimisation problems: predicting optimal depth-of-win for the KRK endgame, and jobshop scheduling. Our results are promising: (i)On each iteration of distribution estimation, samples obtained with an ILP-assisted DBN have a substantially greater proportion of good solutions than samples generated using a DBN without ILP features, and (ii)On termination of distribution estimation, samples obtained using an ILP-assisted DBN contain more near-optimal samples than samples from a DBN without ILP features. These results suggest that the use of ILP-constructed theories could be useful for incorporating complex domain-knowledge into deep models for estimation of distribution based procedures.
no code implementations • 22 Aug 2016 • Pankaj Malhotra, Vishnu Tv, Anusha Ramakrishnan, Gaurangi Anand, Lovekesh Vig, Puneet Agarwal, Gautam Shroff
Many approaches for estimation of Remaining Useful Life (RUL) of a machine, using its operational sensor data, make assumptions about how a system degrades or a fault evolves, e. g., exponential degradation.
no code implementations • 3 Aug 2016 • Ashwin Srinivasan, Gautam Shroff, Lovekesh Vig, Sarmimala Saikia, Puneet Agarwal
To answer this in the affirmative, we need: (a)a general-purpose technique for the incorporation of domain knowledge when constructing models for optimal values; and (b)a way of using these models to generate new data samples.
9 code implementations • 1 Jul 2016 • Pankaj Malhotra, Anusha Ramakrishnan, Gaurangi Anand, Lovekesh Vig, Puneet Agarwal, Gautam Shroff
Mechanical devices such as engines, vehicles, aircrafts, etc., are typically instrumented with numerous sensors to capture the behavior and health of the machine.
no code implementations • 5 May 2016 • Mohit Yadav, Pankaj Malhotra, Lovekesh Vig, K Sriram, Gautam Shroff
The available data is then augmented with data generated from the ODE, and the anomaly detector is retrained on this augmented dataset.
1 code implementation • ESANN 2015 • Pankaj Malhotra, Lovekesh Vig, Gautam Shroff, Puneet Agarwal
Long Short Term Memory (LSTM) networks have been demonstrated to be particularly useful for learning sequences containing longer term patterns of unknown length, due to their ability to maintain long term memory.