125 citations · 162 across the 12 of their papers we have counts for
13 papers
Jointly Reparametrized Multi-Layer Adaptation for Efficient and Private Tuning
Umang Gupta, Aram Galstyan, Greg Ver Steeg
Efficient finetuning of pretrained language transformers is becoming increasingly prevalent for solving natural language processing tasks. While effective, it can still require a l…
Transferring Models Trained on Natural Images to 3D MRI via Position Encoded Slice Models
Umang Gupta, Tamoghna Chattopadhyay, Nikhil Dhinagar +3
Transfer learning has remarkably improved computer vision. These advances also promise improvements in neuroimaging, where training set sizes are often small. However, various diff…
Towards Sparsified Federated Neuroimaging Models via Weight Pruning
Dimitris Stripelis, Umang Gupta, Nikhil Dhinagar +3
Federated training of large deep neural networks can often be restrictive due to the increasing costs of communicating the updates with increasing model sizes. Various model prunin…
Secure & Private Federated Neuroimaging
Dimitris Stripelis, Umang Gupta, Hamza Saleem +9
The amount of biomedical data continues to grow rapidly. However, collecting data from multiple sites for joint analysis remains challenging due to security, privacy, and regulator…
Federated Progressive Sparsification (Purge, Merge, Tune)+
Dimitris Stripelis, Umang Gupta, Greg Ver Steeg +1
To improve federated training of neural networks, we develop FedSparsify, a sparsification strategy based on progressive weight magnitude pruning. Our method has several benefits.…
Mitigating Gender Bias in Distilled Language Models via Counterfactual Role Reversal
Umang Gupta, Jwala Dhamala, Varun Kumar +7
Language models excel at generating coherent text, and model compression techniques such as knowledge distillation have enabled their use in resource-constrained settings. However,…
Attributing Fair Decisions with Attention Interventions
Ninareh Mehrabi, Umang Gupta, Fred Morstatter +2
The widespread use of Artificial Intelligence (AI) in consequential domains, such as healthcare and parole decision-making systems, has drawn intense scrutiny on the fairness of th…
Secure Neuroimaging Analysis using Federated Learning with Homomorphic Encryption
Dimitris Stripelis, Hamza Saleem, Tanmay Ghai +8
Federated learning (FL) enables distributed computation of machine learning models over various disparate, remote data sources, without requiring to transfer any individual data to…
Membership Inference Attacks on Deep Regression Models for Neuroimaging
Umang Gupta, Dimitris Stripelis, Pradeep K. Lam +3
Ensuring the privacy of research participants is vital, even more so in healthcare environments. Deep learning approaches to neuroimaging require large datasets, and this often nec…
Improved Brain Age Estimation with Slice-based Set Networks
Umang Gupta, Pradeep K. Lam, Greg Ver Steeg +1
Deep Learning for neuroimaging data is a promising but challenging direction. The high dimensionality of 3D MRI scans makes this endeavor compute and data-intensive. Most conventio…
Controllable Guarantees for Fair Outcomes via Contrastive Information Estimation
Umang Gupta, Aaron M Ferber, Bistra Dilkina +1
Controlling bias in training datasets is vital for ensuring equal treatment, or parity, between different groups in downstream applications. A naive solution is to transform the da…
Deep Generative Dual Memory Network for Continual Learning
Nitin Kamra, Umang Gupta, Yan Liu
Despite advances in deep learning, neural networks can only learn multiple tasks when trained on them jointly. When tasks arrive sequentially, they lose performance on previously l…
Automated determination of g-mode period spacing of red-giant stars
Abhisek Datta, Anwesh Mazumdar, Umang Gupta +1
The Kepler satellite has provided photometric timeseries data of unprecedented length, duty cycle and precision. To fully analyse these data for the tens of thousands of stars obse…