AI summary

This paper introduces memory networks, a new class of learning models that reasons with inference components combined with a long-term memory component. The models are evaluated on question answering tasks, where the long-term memory acts as a dynamic knowledge base. The models are shown to have reasoning capabilities.

Abstract

We describe a new class of learning models called memory networks. Memory networks reason with inference components combined with a long-term memory component; they learn how to use these jointly. The long-term memory can be read and written to, with the goal of using it for prediction. We investigate these models in the context of question answering (QA) where the long-term mem ory effectively acts as a (dynamic) knowledge base, and the output is a textual response. We evaluate them on a large-scale QA task, and a smaller, but more complex, toy task generated from a simulated world. In the latter, we show the reasoning power of such models by chaining multiple supporting sentences to an swer questions that require understanding the intension of verbs.