Presentation | 2021-03-03 Safe reinforcement learning in high-dimensional continuous spaces Takumi Umemoto, Tohgoroh Matsui, Atsuko Mutoh, Koich Moriyama, Inuzuka Nobuhiro, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | We propose a method to extend the reinforcement learning method (CSEQ) based on success probability and profit in continuous state space to higher dimensions. Reinforcement learning is a machine learning method that learns better behavior based on trial and error, and there is a method called EQ based on success probability and profit as safe reinforcement learning focusing on learning danger avoidance behavior, and continuous observation Its effectiveness has been confirmed in the problem on the two-dimensional space of. We propose safe reinforcement learning that deals with high-dimensional continuous space using the mean values of latent variables modeled by VAE. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | reinforcement learning / safe reinforcement learning / deep learning / auto encoder |
Paper # | IBISML2020-50 |
Date of Issue | 2021-02-23 (IBISML) |
Conference Information | |
Committee | IBISML |
---|---|
Conference Date | 2021/3/2(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Organized and general sessions on machine learning |
Chair | Ichiro Takeuchi(Nagoya Inst. of Tech.) |
Vice Chair | Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo) |
Secretary | Masashi Sugiyama(AIST) / Koji Tsuda(NTT) |
Assistant | Atsuyoshi Nakamura(Hokkaido Univ.) / Shigeyuki Oba(Miidas) |
Paper Information | |
Registration To | Technical Committee on Infomation-Based Induction Sciences and Machine Learning |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Safe reinforcement learning in high-dimensional continuous spaces |
Sub Title (in English) | |
Keyword(1) | reinforcement learning |
Keyword(2) | safe reinforcement learning |
Keyword(3) | deep learning |
Keyword(4) | auto encoder |
1st Author's Name | Takumi Umemoto |
1st Author's Affiliation | Nagoya Institute of Technology(NIT) |
2nd Author's Name | Tohgoroh Matsui |
2nd Author's Affiliation | Chubu University(Chubu Univ.) |
3rd Author's Name | Atsuko Mutoh |
3rd Author's Affiliation | Nagoya Institute of Technology(NIT) |
4th Author's Name | Koich Moriyama |
4th Author's Affiliation | Nagoya Institute of Technology(NIT) |
5th Author's Name | Inuzuka Nobuhiro |
5th Author's Affiliation | Nagoya Institute of Technology(NIT) |
Date | 2021-03-03 |
Paper # | IBISML2020-50 |
Volume (vol) | vol.120 |
Number (no) | IBISML-395 |
Page | pp.pp.55-62(IBISML), |
#Pages | 8 |
Date of Issue | 2021-02-23 (IBISML) |