Shimon Whiteson

A model-free method for decision-making in uncertain environments using Bayesian techniques.

Research shows how simple models outperform complex methods in Meta-RL tasks.

Researchers develop a method for efficient robot control across different designs.

Exploring new methods to enhance decision-making in learning agents.

A new method improves training data for offline reinforcement learning.

A new method helps agents learn from experts despite missing information.

A novel method addresses key challenges in reinforcement learning through improved optimization techniques.

Researchers tackle challenges of teaching self-driving cars through imitation and learning.

New techniques improve learning efficiency in AI agents as environments shift.