Dialog Policy Learning for Joint Clarification and Active Learning Queries
This work addresses the need for dialog systems to handle uncertainty and learn new concepts in real-time, which is incremental as it combines existing functions into a unified policy.
The paper tackles the problem of enabling intelligent systems to recover from mistakes and adapt to novel concepts by training a hierarchical dialog policy that jointly performs clarification and active learning queries in an interactive language-based image retrieval task, demonstrating that this joint approach is more effective than using static policies for these functions.
Intelligent systems need to be able to recover from mistakes, resolve uncertainty, and adapt to novel concepts not seen during training. Dialog interaction can enable this by the use of clarifications for correction and resolving uncertainty, and active learning queries to learn new concepts encountered during operation. Prior work on dialog systems has either focused on exclusively learning how to perform clarification/ information seeking, or to perform active learning. In this work, we train a hierarchical dialog policy to jointly perform both clarification and active learning in the context of an interactive language-based image retrieval task motivated by an online shopping application, and demonstrate that jointly learning dialog policies for clarification and active learning is more effective than the use of static dialog policies for one or both of these functions.