CoChat

Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells

Open access
Nekrasov, Vladimir, Chen, Hao, Shen, Chunhua, Reid, Ian · 2019 · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Cited by 157

The DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.

Abstract

Automated design of neural network architectures tailored for a specific task is an extremely promising, albeit inherently difficult, avenue to explore. While most results in this domain have been achieved on image classification and language modelling problems, here we concentrate on dense per-pixel tasks, in particular, semantic image segmentation using fully convolutional networks. In contrast to the aforementioned areas, the design choices of a fully convolutional network require several changes, ranging from the sort of operations that need to be used - e.g., dilated convolutions - to a solving of a more difficult optimisation problem. In this work, we are particularly interested in searching for high-performance compact segmentation architectures, able to run in real-time using limited resources. To achieve that, we intentionally over-parameterise the architecture during the training time via a set of auxiliary cells that provide an intermediate supervisory signal and can be omitted during the evaluation phase. The design of the auxiliary cell is emitted by a controller, a neural network with the fixed structure trained using reinforcement learning. More crucially, we demonstrate how to efficiently search for these architectures within limited time and computational budgets. In particular, we rely on a progressive strategy that terminates non-promising architectures from being further trained, and on Polyak averaging coupled with knowledge distillation to speed-up the convergence. Quantitatively, in 8 GPU-days our approach discovers a set of architectures performing on-par with state-of-the-art among compact models on the semantic segmentation, pose estimation and depth prediction tasks. Code will be made available here: https://github.com/drsleep/nas-segm-pytorch.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr.2019.00934

Cites

22 of 22 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 22 references

Research works printed but not linked: 27

  1. 10.1162/artl.2009.15.2.15202 · DOI · Paper page
  2. 10.1109/IROS.2018.8594243 · DOI
  3. 10.1109/CVPR.2013.471 · DOI · Paper page
  4. sandler, 2018, Inverted residuals and linear bottlenecks: Mobile networks for classification, detection and segmentation, Proc IEEE Conf Comp Vis Patt Recogn
  5. real, 2017, Large-scale evolution of image classifiers, Proc Int Conf Mach Learn
  6. glorot, 2010, Understanding the difficulty of training deep feedforward neural networks, Proc Conf Artif Intell Statist
  7. 10.1162/106365602320169811 · DOI · Paper page
  8. 10.1109/ICCV.2011.6126343 · DOI · Paper page
  9. hinton, 2014, Distilling the knowledge in a neural network, Proc Advances in Neural Inf Process Syst
  10. kingma, 2014, Adam: A method for stochastic optimization, ArXiv Computing Research Repository
  11. lin, 2017, Exploring context with deep structured models for semantic segmentation, IEEE Trans Pattern Anal Mach Intell
  12. pham, 2018, Efficient neural architecture search via parameter sharing, Proc Int Conf Mach Learn
  13. bergstra, 2013, Making a science of model search: Hyperparameter optimization in hundreds of dimensions for vision architectures, Proc Int Conf Mach Learn
  14. baker, 2017, Designing neural network architectures using reinforcement learning, Proc Int Conf Learn Representations
  15. 10.1109/TPAMI.2017.2699184 · DOI · Paper page
  16. du, 2018, On the power of over-parametrization in neural networks with quadratic activation, Proc Int Conf Mach Learn
  17. 10.1109/72.265960 · DOI · Paper page
  18. 10.1007/s11263-009-0275-4 · DOI · Paper page
  19. 10.1109/CVPR.2014.471 · DOI · Paper page
  20. lin, 2014, Microsoft COCO: common objects in context, Proc Eur Conf Comp Vis
  21. liu, 2018, Hierarchical representations for efficient architecture search, Proc Int Conf Learn Representations
  22. watkins, 1989, Learning from delayed rewards, PhD thesis
  23. luo, 2018, Neural architecture opti-mization, Proc Advances in Neural Inf Process Syst
  24. 10.1109/CVPR.2015.7298965 · DOI · Paper page
  25. 10.1109/CVPR.2017.75 · DOI · Paper page
  26. nekrasov, 2018, Light-weight refinenet for real-time semantic segmentation, Proc British Machine Vis Conf
  27. xiao, 2018, Simple baselines for human pose estimation and tracking, Proc Eur Conf Comp Vis

Cited by

Metadata

Authors
Nekrasov, V., Chen, H., Shen, C., & Reid, I.
Venue
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2019-06
Pages
9118-9127
ISSN
1063-6919 (ISSN-L)
OpenAlex
W2898245822
Cited by
157
References
22
FWCI
9.78
Type
conference paper
Language
en
Open access
green

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.