Neural Approaches to Conversational AI

Conversational AI is fundamental to natural user interfaces. It is a rapidly growing field, attracting many researchers in the Natural Language Processing (NLP), Information Retrieval (IR) and Machine Learning (ML) communities. For example, SIGIR 2018 has created a new track of Artificial Intelligence, Semantics, and Dialog to bridge research in AI and IR, especially targeting Question Answering (QA), deep semantics and dialogue with intelligent agents.

·      We provide a comprehensive survey of the neural approaches to conversational AI that have been developed in the last few years, covering QA, task-oriented and social bots with a unified view of optimal decision making.

·      We draw connections between modern neural approaches and traditional approaches, allowing us to better understand why and how the research has evolved and to shed light on how we can move forward.

·      We present state-of-the-art approaches to training dialogue agents using both supervised and reinforcement learning.