This is not the latest version of this item. The latest version can be found here.
Neurally-guided structure inference
Name
2019ICML-NGSI.pdf
Description
Published version
Size
1012.38 KB
Format
Adobe PDF
Checksum (MD5)
42dece6c5d11690042b9946ea1c1d37e
Author(s)
Lu, S
Mao, J
Tenenbaum, JB
Wu, J
Date Issued
January 1, 2019
Journal
36th International Conference on Machine Learning, ICML 2019
Citation
Lu, S, Mao, J, Tenenbaum, JB and Wu, J. 2019. "Neurally-guided structure inference." 36th International Conference on Machine Learning, ICML 2019, 2019-June.
Version
Final published version
Abstract
© 36th International Conference on Machine Learning, ICML 2019. All rights reserved. Most structure inference methods either rely on exhaustive search or are purely data-driven. Exhaustive search robustly infers the structure of arbitrarily complex data, but it is slow. Data-driven methods allow efficient inference, but do not generalize when test data have more complex structures than training data. In this paper, we propose a hybrid inference algorithm, the Neurally-Guided Structure Inference (NG-SI), keeping the advantages of both search-based and data-driven methods. The key idea of NG-SI is to use a neural network to guide the hierarchical, layer-wise search over the compositional space of structures. We evaluate our algorithm on two representative structure inference tasks: probabilistic matrix decomposition and symbolic program parsing. It outperforms data-driven and search-based alternatives on both tasks.
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
http://ngsi.csail.mit.edu/data/papers/2019ICML-NGSI.pdf
