|
JAIST Repository >
School of Transdisciplinary Science >
Conference Papers >
Conference Papers >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10119/18718
|
| Title: | Improving the Human-Likeness of Game AI’s Moves by Combining Multiple Prediction Models |
| Authors: | Ogawa, Tatsuyoshi
Hsueh, Chu-Hsuan
Ikeda, Kokolo |
| Keywords: | Human-Likeness
Player Modeling
Move Prediction
AlphaZero
Shogi |
| Issue Date: | 2023-02 |
| Publisher: | SCITEPRESS – Science and Technology Publications, Lda. |
| Magazine name: | Proceedings of the 15th International Conference on Agents and Artificial Intelligence |
| Volume: | 3 |
| Start page: | 931 |
| End page: | 939 |
| DOI: | 10.5220/0011804200003393 |
| Abstract: | Strong game AI’s moves are sometimes strange or difficult for humans to understand. To achieve better human-computer interaction, researchers try to create human-like game AI. For chess and Go, supervised learning with deep neural networks is one of the most effective methods to predict human moves. In this study, we first show that supervised learning is also effective in Shogi (Japanese chess) to predict human moves. We also find that the AlphaZero-based model more accurately predicted moves of players with higher skill. We then investigate two evaluation metrics for measuring human-likeness, where one is move-matching accuracy that is often used in existing works, and the other is likelihood (the geometric mean of human moves’ probabilities predicted by the model). To create game AI that is more human-like, we propose two methods to combine multiple move prediction models. One uses a Classifier to select a suitable prediction model according to different situations, and the other is Blend that mixes probabilities from different prediction models because we observe that each model is good at some situations where other models cannot predict well. We show that the Classifier method increases the move-matching accuracy by 1%-3% but fails to improve the likelihood. The Blend method increases the move-matching accuracy by 3%-4% and the likelihood by 2%-5%. |
| Rights: | Copyright (C) 2023 SCITEPRESS - Science and Technology Publications. Tatsuyoshi Ogawa, Chu-Hsuan Hsueh, Kokolo Ikeda, Proceedings of the 15th International Conference on Agents and Artificial Intelligence (ICAART 2023), 3, 2023, 931-939. DOI: 10.5220/0011804200003393. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. |
| URI: | http://hdl.handle.net/10119/18718 |
| Material Type: | publisher |
| Appears in Collections: | d11-1. 会議発表論文 (Conference Papers)
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| I-IKEDA-K0405-19.pdf | | 1730Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
