Interactive Machine Comprehension with Dynamic Knowledge Graphs

TL;DR

This paper introduces a method for interactive machine reading comprehension that uses dynamic knowledge graphs as memory, enabling agents to better gather and utilize information for answering questions, with significant performance gains demonstrated on iSQuAD.

Contribution

It proposes a novel approach integrating dynamic graph representations into RL agents for iMRC, enhancing their ability to reason with partially observable knowledge.

Findings

Graph representations improve RL agent performance in iMRC.

Dynamic graph building and updating are effective for information gathering.

Significant performance improvements demonstrated on iSQuAD.

Abstract

Interactive machine reading comprehension (iMRC) is machine comprehension tasks where knowledge sources are partially observable. An agent must interact with an environment sequentially to gather necessary knowledge in order to answer a question. We hypothesize that graph representations are good inductive biases, which can serve as an agent's memory mechanism in iMRC tasks. We explore four different categories of graphs that can capture text information at various levels. We describe methods that dynamically build and update these graphs during information gathering, as well as neural models to encode graph representations in RL agents. Extensive experiments on iSQuAD suggest that graph representations can result in significant performance improvements for RL agents.