Can I read Zero-Shot Reinforcement Learning via Function Encoders on EtoBox?

Zero-Shot Reinforcement Learning via Function Encoders by Ingebrand, Tyler; Zhang, Amy; Topcu, Ufuk is a scholarly article available to read on EtoBox.

What is Zero-Shot Reinforcement Learning via Function Encoders about?

Although reinforcement learning (RL) can solve many challenging sequential decision making problems, achieving zero-shot transfer across related tasks remains a challenge. The difficulty lies in finding a good representation for the current task so that the agent understands how it relates to previously seen tasks. To achieve zero-shot transfer, we introduce the function encoder, a representation learning algorithm which represents a function as a weighted combination of learned, non-linear basis functions. By using a function encoder to represent the reward function or the transition function, the agent has information on how the current task relates to previously seen tasks via a coherent vector representation. Thus, the agent is able to achieve transfer between related tasks at run time with no additional training. We demonstrate state-of-the-art data efficiency, asymptotic performance, and training stability in three RL fields by augmenting basic RL algorithms with a function encoder task representation.

Author

Ingebrand, Tyler; Zhang, Amy; Topcu, Ufuk

Published

2024

Language

EN