Learning Open Domain Multi-hop Search Using Reinforcement Learning
This addresses the challenge of efficient information retrieval for multi-hop queries, but it is incremental as it builds on existing methods with specific improvements.
The paper tackles the problem of teaching an automated agent to search for multi-hop paths between entities in an open domain, resulting in policies that process fewer documents than baseline heuristic algorithms.
We propose a method to teach an automated agent to learn how to search for multi-hop paths of relations between entities in an open domain. The method learns a policy for directing existing information retrieval and machine reading resources to focus on relevant regions of a corpus. The approach formulates the learning problem as a Markov decision process with a state representation that encodes the dynamics of the search process and a reward structure that minimizes the number of documents that must be processed while still finding multi-hop paths. We implement the method in an actor-critic reinforcement learning algorithm and evaluate it on a dataset of search problems derived from a subset of English Wikipedia. The algorithm finds a family of policies that succeeds in extracting the desired information while processing fewer documents compared to several baseline heuristic algorithms.