Probing the compositionality of intuitive functions
Name
6130-probing-the-compositionality-of-intuitive-functions.pdf
Size
510.16 KB
Format
Adobe PDF
Checksum (MD5)
984bfc0a624f832d304f9982946fdc4c
Author(s) • • • •
Schulz, Eric
Duvenaud, David K.
Speekenbrink, Maarten
Gershman, Samuel J.
Tenenbaum, Joshua B
Date Issued
December 2016
Journal
30th Conference on Neural Information Processing Systems (NIPS 2016)
Publisher
Neural Information Processing Systems Foundation
Citation
Schulz, Eric et al. "Probing the Compositionality of Intuitive Functions." Advances in Neural Information Processing Systems 29 (NIPS 2016), Barcelona, Spain, December 5-10, 2016. © 2016 Neural Information Processing Systems Foundation
Version
Final published version
Abstract
How do people learn about complex functional structure? Taking inspiration from other areas of cognitive science, we propose that this is accomplished by harnessing compositionality: complex structure is decomposed into simpler building blocks. We formalize this idea within the framework of Bayesian regression using a grammar over Gaussian process kernels. We show that participants prefer compositional over non-compositional function extrapolations, that samples from the human prior over functions are best described by a compositional model, and that people perceive compositional functions as more predictable than their non-compositional but otherwise similar counterparts. We argue that the compositional nature of intuitive
functions is consistent with broad principles of human cognition.
MIT Department
Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences
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
https://papers.nips.cc/paper/6130-probing-the-compositionality-of-intuitive-functions