Publications (9)
On the Limits of Multi-modal Meta-Learning with Auxiliary Task Modulation Using Conditional Batch Normalization
Jordi Armengol-Estapé, Vincent Michalski, Ramnath Kumar +3
Few-shot learning aims to learn representations that can tackle novel tasks given a small number of examples. Recent studies show that cross-modal learning can improve representati…
FlexAct: Why Learn when you can Pick?
Ramnath Kumar, Kyle Ritscher, Junmin Judy +2
Learning activation functions has emerged as a promising direction in deep learning, allowing networks to adapt activation mechanisms to task-specific demands. In this work, we int…
EHI: End-to-end Learning of Hierarchical Index for Efficient Dense Retrieval
Ramnath Kumar, Anshul Mittal, Nilesh Gupta +3
Dense embedding-based retrieval is widely used for semantic search and ranking. However, conventional two-stage approaches, involving contrastive embedding learning followed by app…
FastLane: Efficient Routed Systems for Late-Interaction Retrieval
Ramnath Kumar, Prateek Jain, Cho-Jui Hsieh
Late-interaction retrieval models like ColBERT achieve superior accuracy by enabling token-level interactions, but their computational cost hinders scalability and integration with…
Introspective Experience Replay: Look Back When Surprised
Ramnath Kumar, Dheeraj Nagaraj
In reinforcement learning (RL), experience replay-based sampling techniques play a crucial role in promoting convergence by eliminating spurious correlations. However, widely used…
Stochastic Re-weighted Gradient Descent via Distributionally Robust Optimization
Ramnath Kumar, Kushal Majmundar, Dheeraj Nagaraj +1
We present Re-weighted Gradient Descent (RGD), a novel optimization technique that improves the performance of deep neural networks through dynamic sample re-weighting. Leveraging…
Boosting Exploration in Multi-Task Reinforcement Learning using Adversarial Networks
Ramnath Kumar, Tristan Deleu, Yoshua Bengio
Advancements in reinforcement learning (RL) have been remarkable in recent years. However, the limitations of traditional training methods have become increasingly evident, particu…
MIRAGE: Protecting against Malicious Image Editing via False Moderation
Anshul Nasery, Ramnath Kumar, Cho-Jui Hsieh +1
The proliferation of AI-powered image editing systems raises serious concerns because it allows personal images to be arbitrarily manipulated at scale, with minimal effort, and a l…
The Effect of Diversity in Meta-Learning
Ramnath Kumar, Tristan Deleu, Yoshua Bengio
Recent studies show that task distribution plays a vital role in the meta-learner's performance. Conventional wisdom is that task diversity should improve the performance of meta-l…