Modeling question asking using neural program generation

Ziyun Wang, Brenden M. Lake

Keywords: question asking, language generation, program induction, reinforcement learning, density estimation, cognitive science

TL;DR: We introduce a model of human question asking that combines neural networks and symbolic programs, which can learn to generate good questions with or without supervised examples.

Abstract: People ask questions that are far richer, more informative, and more creative than current AI systems. We propose a neural program generation framework for modeling human question asking, which represents questions as formal programs and generates programs with an encoder-decoder based deep neural network. From extensive experiments using an information-search game, we show that our method can ask optimal questions in synthetic settings, and predict which questions humans are likely to ask in unconstrained settings. We also propose a novel grammar-based question generation framework trained with reinforcement learning, which is able to generate creative questions without supervised data.