How Do We Teach Reinforcement Learning Agents Human Preferences?

Table of Links

1. Abstract and Introduction
2. Related Work
3. Problem Definition
4. Method
5. Experiments
6. Conclusion and References

A. Appendix

A.1. Full Prompts and A.2 ICPL Details

A. 3 Baseline Details

A.4 Environment Details

A.5 Proxy Human Preference

A.6 Human-in-the-Loop Preference

3 PROBLEM DEFINITION

Our goal is to design a reward function that can be used to train reinforcement learning agents that demonstrate human-preferred behaviors. It is usually hard to design proper reward functions in reinforcement learning that induce policies that align well with human preferences.

:::info
Authors:

(1) Chao Yu, Tsinghua University;

(2) Hong Lu, Tsinghua University;

(3) Jiaxuan Gao, Tsinghua University;

(4) Qixin Tan, Tsinghua University;

(5) Xinting Yang, Tsinghua University;

(6) Yu Wang, with equal advising from Tsinghua University;

(7) Yi Wu, with equal advising from Tsinghua University and the Shanghai Qi Zhi Institute;

(8) Eugene Vinitsky, with equal advising from New York University (zoeyuchao@gmail.com).

:::

---

:::info
This paper is available on arxiv under CC 4.0 license.

:::