Differentiable Programs with Neural Libraries

We develop a framework for combining differentiable programming languages with neural networks. Using this framework we create end-to-end trainable systems that learn to write interpretable algorithms with perceptual components. We explore the benefits of inductive biases for strong generalization a...

Full description

Saved in:
Bibliographic Details
Main Authors Gaunt, Alexander L, Brockschmidt, Marc, Kushman, Nate, Tarlow, Daniel
Format Journal Article
LanguageEnglish
Published 07.11.2016
Subjects
Computer Science - Learning
Online AccessGet full text

Cover

More Information
Summary:We develop a framework for combining differentiable programming languages with neural networks. Using this framework we create end-to-end trainable systems that learn to write interpretable algorithms with perceptual components. We explore the benefits of inductive biases for strong generalization and modularity that come from the program-like structure of our models. In particular, modularity allows us to learn a library of (neural) functions which grows and improves as more tasks are solved. Empirically, we show that this leads to lifelong learning systems that transfer knowledge to new tasks more effectively than baselines.
DOI:10.48550/arxiv.1611.02109