11 papers
Richer Bayesian Last Layers with Subsampled NTK Features
Sergio Calvo-Ordoñez, Jonathan Plenk, Richard Bergna +4
Bayesian Last Layers (BLLs) provide a convenient and computationally efficient way to estimate uncertainty in neural networks. However, they underestimate epistemic uncertainty bec…
Stabilizing Policy Gradients for Sample-Efficient Reinforcement Learning in LLM Reasoning
Luckeciano C. Melo, Alessandro Abate, Yarin Gal
Reinforcement Learning, particularly through policy gradient methods, has played a central role in enabling reasoning capabilities of Large Language Models. However, the optimizati…
Weighted Conditional Flow Matching
Sergio Calvo-Ordonez, Matthieu Meunier, Alvaro Cartea +3
Conditional flow matching (CFM) has emerged as a powerful framework for training continuous normalizing flows due to its computational efficiency and effectiveness. However, standa…
Iterative Deployment Improves Planning Skills in LLMs
Augusto B. Corrêa, Yoav Gelberg, Luckeciano C. Melo +3
We show that iterative deployment of large language models (LLMs), each fine-tuned on data carefully curated by users from the previous models' deployment, can significantly change…
Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning
Gunshi Gupta, Karmesh Yadav, Zsolt Kira +2
To enable embodied agents to operate effectively over extended timeframes, it is crucial to develop models that form and access memories to stay contextualized in their environment…
Temporal-Difference Variational Continual Learning
Luckeciano C. Melo, Alessandro Abate, Yarin Gal
Machine Learning models in real-world applications must continuously learn new tasks to adapt to shifts in the data-generating distribution. Yet, for Continual Learning (CL), model…