|
JAIST Repository >
d. 融合科学系 >
d11. 会議発表論文 >
d11-1. 会議発表論文 >
このアイテムの引用には次の識別子を使用してください:
http://hdl.handle.net/10119/18239
|
| タイトル: | Imitating Agents in A Complex Environment by Generative Adversarial Imitation Learning |
| 著者: | Li, Wanxiang
Hsueh, Chu-Hsuan
Ikeda, Kokolo |
| キーワード: | Reinforcement Learning
Generative Adversarial Networks
Super Mario Bros
Generative Adversarial Imitation Learning |
| 発行日: | 2020-10-20 |
| 出版者: | Institute of Electrical and Electronics Engineers (IEEE) |
| 誌名: | 2020 IEEE Conference on Games (CoG 2020) |
| 巻: | 2020 |
| DOI: | 10.1109/CoG47356.2020.9231805 |
| 抄録: | The generative adversarial imitation learning (GAIL) shows the ability to find reward functions to explain expert players’ behaviors in some low-dimensional environments using hand-crafted features as inputs. In this research, we aim to extend GAIL to complex environments and using raw images as inputs. We propose to (1) use convolutional neuron networks to deal with image inputs, (2) adopt a structure called globallocal discriminator to GAIL, and (3) represent trajectories as state-state pairs instead of state-action pairs. Our approach successfully imitates given players in Super Mario Bros. To our knowledge, the results are the first to have successful imitations in complex environments based on image inputs. |
| Rights: | This is the author's version of the work. Copyright (C) 2020 IEEE. 2020 IEEE Conference on Games (CoG). DOI: 10.1109/CoG47356.2020.9231805. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. |
| URI: | http://hdl.handle.net/10119/18239 |
| 資料タイプ: | author |
| 出現コレクション: | d11-1. 会議発表論文 (Conference Papers)
|
このアイテムのファイル:
| ファイル |
記述 |
サイズ | 形式 |
|---|
| I-IKEDA-K0405-15.pdf | | 248Kb | Adobe PDF | 見る/開く |
|
当システムに保管されているアイテムはすべて著作権により保護されています。
|
