Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks
This work addresses the problem of evaluating AI systems for intelligent dialogue, particularly for researchers aiming to build human-like conversational agents, but it is incremental as it builds on existing proxy task frameworks.
The paper tackles the challenge of measuring progress towards AI-complete question answering by proposing a set of prerequisite toy tasks that evaluate reading comprehension through various reasoning skills like chaining facts and deduction. They extend the Memory Networks model, showing it can solve some but not all tasks, with no concrete performance numbers provided.
One long-term goal of machine learning research is to produce methods that are applicable to reasoning and natural language, in particular building an intelligent dialogue agent. To measure progress towards that goal, we argue for the usefulness of a set of proxy tasks that evaluate reading comprehension via question answering. Our tasks measure understanding in several ways: whether a system is able to answer questions via chaining facts, simple induction, deduction and many more. The tasks are designed to be prerequisites for any system that aims to be capable of conversing with a human. We believe many existing learning systems can currently not solve them, and hence our aim is to classify these tasks into skill sets, so that researchers can identify (and then rectify) the failings of their systems. We also extend and improve the recently introduced Memory Networks model, and show it is able to solve some, but not all, of the tasks.