Large-scale Simple Question Answering with Memory Networks

TL;DR

This paper explores how multitask and transfer learning can improve large-scale simple question answering systems using Memory Networks, introducing a new dataset and demonstrating successful training for high performance.

Contribution

It introduces a new large-scale dataset for simple question answering and demonstrates the effectiveness of multitask and transfer learning within Memory Networks.

Findings

Memory Networks can be effectively trained for large-scale QA.

Multitask and transfer learning improve performance.

A new dataset of 100k questions supports large-scale evaluation.

Abstract

Training large-scale question answering systems is complicated because training sources usually cover a small portion of the range of possible questions. This paper studies the impact of multitask and transfer learning for simple question answering; a setting for which the reasoning required to answer is quite easy, as long as one can retrieve the correct evidence given a question, which can be difficult in large-scale conditions. To this end, we introduce a new dataset of 100k questions that we use in conjunction with existing benchmarks. We conduct our study within the framework of Memory Networks (Weston et al., 2015) because this perspective allows us to eventually scale up to more complex reasoning, and show that Memory Networks can be successfully trained to achieve excellent performance.