Learning to Compose Neural Networks for Question Answering release_jrrrrp5v5nab5hpium5t3tazki

by Jacob Andreas, Marcus Rohrbach, Trevor Darrell, Dan Klein

Released as a article .

2016  

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.
In text/plain format

Archived Files and Locations

application/pdf  3.2 MB
file_hxcdy74zyfh6fjiag7w5bjfxje
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2016-06-01
Version   v2
Language   en ?
arXiv  1601.01705v2
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: d373a97e-3570-4288-bc7e-300d8c84d08c
API URL: JSON