4 papers
Boosting Accuracy and Efficiency of Budget Forcing in LLMs via Reinforcement Learning for Mathematical Reasoning
Ravindra Aribowo Tarunokusumo, Rafael Fernandes Cunha
Test-time scaling methods have seen a rapid increase in popularity for its computational efficiency and parameter-independent training to improve reasoning performance on Large Lan…
Sparsity-Driven Plasticity in Multi-Task Reinforcement Learning
Aleksandar Todorov, Juan Cardenas-Cartagena, Rafael F. Cunha +2
Plasticity loss, a diminishing capacity to adapt as training progresses, is a critical challenge in deep reinforcement learning. We examine this issue in multi-task reinforcement l…
World Model Agents with Change-Based Intrinsic Motivation
Jeremias Ferrao, Rafael Cunha
Sparse reward environments pose a significant challenge for reinforcement learning due to the scarcity of feedback. Intrinsic motivation and transfer learning have emerged as promi…
Optimally Solving Simultaneous-Move Dec-POMDPs: The Sequential Central Planning Approach
Johan Peralez, Aurèlien Delage, Jacopo Castellini +2
The centralized training for decentralized execution paradigm emerged as the state-of-the-art approach to -optimally solving decentralized partially observable Markov decision…