Reinforcement Learning in Robotics

This paper shows that selectively maintaining high-value information helps resolve the exploration/exploitation dilemma in decision-making, performing as well as uncertainty-driven exploration but with lower cognitive load.

The paper proposes a modular reinforcement learning model to estimate rewards and discount factors from human behavior data, enabling accurate prediction of navigation decisions in virtual reality.

The paper extends a unified account of reinforcement learning to include episodic learning by introducing a meta-learner that coordinates different learning algorithms.

This paper introduces a Bayesian agent that learns to generalize task structures by clustering transition and reward functions independently, showing that performance depends on task domain statistics.

The paper proposes a k-nearest neighbor method to improve parameter estimation in reinforcement learning models, showing better reliability than maximum likelihood approaches.

This paper proposes a framework linking model-based reinforcement learning to model-free mechanisms using successor representations and introduces algorithms that mitigate their limitations.