A Research on Reward Setting and Curiosity Encoding Methods

DOI

10.2991/978-94-6463-300-9_63How to use a DOI?

Keywords

Reward setting; curiosity; reinforcement learning; encoding method

Abstract

Agents in reinforcement learning relies on reward to make movement, improve algorithms, and reach the final goal. However, reward setting is a subject that requires much engineering skills and experiences. Two types of reward, extrinsic reward and intrinsic reward, are totally different in ways of setting. A typical type of intrinsic reward is curiosity. Although there have been many studies on curiosity reward mechanisms in algorithms, the comparison and analysis of different methods are not comprehensive enough. The paper: (a) made detailed introduction to general types of extrinsic reward setting methods and their applications. (b) investigated the encoding methods for curiosity intrinsic reward and make comparisons among various derivations of different type of encoding methods. (c) demonstrated the agent performance implement different encoding methods and prove that encoding method has great influence on the reward setting of curiosity. Finally, the paper summarizes and looks forward to the full text.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Da Yang
PY  - 2023
DA  - 2023/11/27
TI  - A Research on Reward Setting and Curiosity Encoding Methods
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 609
EP  - 617
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_63
DO  - 10.2991/978-94-6463-300-9_63
ID  - Yang2023
ER  -