Learning to Compose Neural Networks for Question Answering

TitleLearning to Compose Neural Networks for Question Answering
Publication TypeJournal Article
Year of Publication2016
AuthorsAndreas, J., Rohrbach M., Darrell T., & Klein D.
Published inCoRR
Volumeabs/1601.01705
Abstract

We describe a question answering model that applies to both images and structured knowledge bases. The model uses natural language strings to automatically assemble neural networks from a collection of composable modules. Parameters for these modules are learned jointly with network-assembly parameters via reinforcement learning, with only (world, question, answer) triples as supervision. Our approach, which we term a dynamic neural model network, achieves state-of-the-art results on benchmark datasets in both visual and structured domains.

URLhttp://www.icsi.berkeley.edu/pubs/vision/neuralnetqanda16.pdf
ICSI Research Group

Vision