Dialog Policy Learning

Dialog policy learning (DPL) aims to create AI agents that can engage in effective and natural conversations by learning optimal strategies for selecting appropriate responses. Current research focuses on improving the efficiency and robustness of reinforcement learning (RL) approaches, often employing transformer-based architectures and incorporating techniques like curriculum learning, adversarial learning, and action embeddings to enhance training and generalization. These advancements are crucial for building more engaging and capable conversational AI systems, impacting fields such as customer service, education, and healthcare through improved user experience and reduced training data requirements.

Papers

January 31, 2024

Scheduled Curiosity-Deep Dyna-Q: Efficient Exploration for Dialog Policy Learning
Xuecheng Niu, Akinori Ito, Takashi Nose
Reinforcement Learning Efficient Exploration Deep Q Task Oriented Dialog Artificial Curiosity Dialog Policy Learning

September 5, 2023

Dialog Action-Aware Transformer for Dialog Policy Learning
Huimin Wang, Wai-Chung Kwan, Kam-Fai Wong
Reinforcement Learning Dialogue Policy Dialog Policy Learning

September 1, 2023

JoTR: A Joint Transformer and Reinforcement Learning Framework for Dialog Policy Learning
Wai-Chung Kwan, Huimin Wang, Hongru Wang, Zezhong Wang, Xian Wu, Yefeng Zheng, Kam-Fai Wong
Dialogue Model Dialogue Policy Reinforcement Learning Framework Dialog Policy Learning Joint Transformer

July 13, 2023

Why Guided Dialog Policy Learning performs well? Understanding the role of adversarial learning and its alternative
Sho Shimoyama, Tetsuro Morimura, Kenshi Abe, Toda Takamichi, Yuta Tomomatsu, Masakazu Sugiyama, Asahi Hentona, Yuuki Azuma, Hirotaka Ninomiya
Reinforcement Learning Integral Role Adversarial Learning Dialogue Policy Dialog Policy Learning

February 27, 2023

Multi-Action Dialog Policy Learning from Logged User Feedback
Shuo Zhang, Junzhou Zhao, Pinghui Wang, Tianxiang Wang, Zi Liang, Jing Tao, Yi Huang, Junlan Feng
Human Feedback Language Reward Dialog Policy Learning

July 1, 2022

Reinforcement Learning of Multi-Domain Dialog Policies Via Action Embeddings
Jorge A. Mendez, Alborz Geramifard, Mohammad Ghavamzadeh, Bing Liu
Reinforcement Learning Action Representation Task Oriented Dialog System Dialogue Domain Dialog Policy Learning

April 12, 2022

Dynamic Dialogue Policy for Continual Reinforcement Learning
Christian Geishauser, Carel van Niekerk, Nurul Lubis, Michael Heck, Hsien-Chin Lin, Shutong Feng, Milica Gašić
Continual LEArning Continual Reinforcement Learning Dialogue Policy Continual Learning Model Dialog Policy Learning