ReasoNet: Learning to Stop Reading in Machine Comprehension

TL;DR

ReasoNet is a neural network architecture for machine comprehension that dynamically determines the number of reasoning turns needed to answer questions, improving performance by learning when to stop reading.

Contribution

It introduces a termination mechanism in neural networks for reading comprehension, enabling adaptive reasoning depth through reinforcement learning.

Findings

Achieved state-of-the-art results on CNN and Daily Mail datasets.

Performed well on Stanford SQuAD and Graph Reachability datasets.

Demonstrated the effectiveness of dynamic reasoning depth.

Abstract

Teaching a computer to read and answer general questions pertaining to a document is a challenging yet unsolved problem. In this paper, we describe a novel neural network architecture called the Reasoning Network (ReasoNet) for machine comprehension tasks. ReasoNets make use of multiple turns to effectively exploit and then reason over the relation among queries, documents, and answers. Different from previous approaches using a fixed number of turns during inference, ReasoNets introduce a termination state to relax this constraint on the reasoning depth. With the use of reinforcement learning, ReasoNets can dynamically determine whether to continue the comprehension process after digesting intermediate results, or to terminate reading when it concludes that existing information is adequate to produce an answer. ReasoNets have achieved exceptional performance in machine comprehension datasets, including unstructured CNN and Daily Mail datasets, the Stanford SQuAD dataset, and a structured Graph Reachability dataset.