Growing a brain: Fine-tuning by increasing model capacity

Yu Xiong Wang, Deva Ramanan, Martial Hebert

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

CNNs have made an undeniable impact on computer vision through the ability to learn high-capacity models with large annotated training sets. One of their remarkable properties is the ability to transfer knowledge from a large source dataset to a (typically smaller) target dataset. This is usually accomplished through fine-tuning a fixed-size network on new target data. Indeed, virtually every contemporary visual recognition system makes use of fine-tuning to transfer knowledge from ImageNet. In this work, we analyze what components and parameters change during finetuning, and discover that increasing model capacity allows for more natural model adaptation through fine-tuning. By making an analogy to developmental learning, we demonstrate that "growing" a CNN with additional units, either by widening existing layers or deepening the overall network, significantly outperforms classic fine-tuning approaches. But in order to properly grow a network, we show that newly-added units must be appropriately normalized to allow for a pace of learning that is consistent with existing units. We empirically validate our approach on several benchmark datasets, producing state-of-the-art results.

Original languageEnglish (US)
Title of host publicationProceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3029-3038
Number of pages10
ISBN (Electronic)9781538604571
DOIs
StatePublished - Nov 6 2017
Externally publishedYes
Event30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 - Honolulu, United States
Duration: Jul 21 2017Jul 26 2017

Publication series

NameProceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
Volume2017-January

Other

Other30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017
Country/TerritoryUnited States
CityHonolulu
Period7/21/177/26/17

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing

Fingerprint

Dive into the research topics of 'Growing a brain: Fine-tuning by increasing model capacity'. Together they form a unique fingerprint.

Cite this