KSC 2017
Current Result Document :
| 한글제목(Korean Title) |
부분 가관측성을 고려하는 Four-rooms 도메인에서의 객체 수집 |
| 영문제목(English Title) |
Gathering Objects in Four-rooms Domain under Partially Observability1 |
| 저자(Author) |
레팜투옌
라엑엠디아부
최승윤
정태충
Le Pham Tuyen
Abu Layek
Seung-yoon Choi
TaeChoong Chung
|
| 원문수록처(Citation) |
VOL 44 NO. 02 PP. 0865 ~ 0867 (2017. 12) |
한글내용
(Korean Abstract) |
Four-rooms domain is a challenging domain for testing hierarchical reinforcement learning algorithm. This paper shows that Deep Recurrent Q-Learning algorithm can achieve a good performance in this domain without using a hierarchical reinforcement learning algorithm. Especially, the algorithm works well in partially observation settings (POMDP).
|
영문내용
(English Abstract) |
|
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
