跳至主導覽
跳至搜尋
跳過主要內容
國立陽明交通大學研發優勢分析平台 首頁
English
中文
在 國立陽明交通大學研發優勢分析平台 搜尋內容
首頁
人員
單位
研究成果
計畫
獎項
活動
貴重儀器
影響
Hierarchical Reinforcement Learning with Guidance for Multi-Domain Dialogue Policy
Mahdin Rohmatillah
,
Jen Tzung Chien
*
*
此作品的通信作者
電機工程學系
研究成果
:
Article
›
同行評審
25
引文 斯高帕斯（Scopus）
總覽
指紋
指紋
深入研究「Hierarchical Reinforcement Learning with Guidance for Multi-Domain Dialogue Policy」主題。共同形成了獨特的指紋。
排序方式
重量
按字母排序
Keyphrases
Hierarchical Reinforcement Learning
100%
Dialogue Policy
100%
Multi-domain Dialogue
100%
High Performance
66%
GPT-2
66%
Language Model
33%
Pruning
33%
Computational Cost
33%
Large-sized
33%
System Evaluation
33%
End System
33%
Low Computation
33%
Human-in-the-loop
33%
Multidomain State
33%
State Representation
33%
Dialogue Systems
33%
Art Performance
33%
Action Representation
33%
Latent Representation
33%
Dialogue Agent
33%
Imitation Learning
33%
Auxiliary Task
33%
Dialogue Management
33%
Action Evaluation
33%
Behavioral Cloning
33%
Multi-domain Dialogue Systems
33%
Simulated Users
33%
Dialogue Policy Learning
33%
Dialogue Policy Optimization
33%
Computer Science
Reinforcement Learning
100%
Dialog System
66%
Experimental Result
33%
Language Modeling
33%
And-States
33%
Computational Cost
33%
System Evaluation
33%
Individual Component
33%
Art Performance
33%
Human-in-the-Loop
33%
Optimization Policy
33%
Dialog Management
33%
Engineering
Reinforcement Learning
100%
Experimental Result
33%
End System
33%
Computational Cost
33%
Individual Component
33%
Chemical Engineering
Reinforcement Learning
100%
Auxiliaries
33%