AM-LFS: AutoML for Loss Function Search

Designing an effective loss function plays an important role in visual analysis. Most existing loss function designs rely on hand-crafted heuristics that require domain experts to explore the large design space, which is usually sub-optimal and time-consuming. In this paper, we propose AutoML for Lo...

Full description

Saved in:
Bibliographic Details
Published in2019 IEEE/CVF International Conference on Computer Vision (ICCV) pp. 8409 - 8418
Main Authors Li, Chuming, Yuan, Xin, Lin, Chen, Guo, Minghao, Wu, Wei, Yan, Junjie, Ouyang, Wanli
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2019
Subjects
Computer vision
Face
Gold
Optimization
Task analysis
Training
Transforms
Online AccessGet full text

Cover

More Information
Summary:Designing an effective loss function plays an important role in visual analysis. Most existing loss function designs rely on hand-crafted heuristics that require domain experts to explore the large design space, which is usually sub-optimal and time-consuming. In this paper, we propose AutoML for Loss Function Search (AM-LFS) which leverages REINFORCE to search loss functions during the training process. The key contribution of this work is the design of search space which can guarantee the generalization and transferability on different vision tasks by including a bunch of existing prevailing loss functions in a unified formulation. We also propose an efficient optimization framework which can dynamically optimize the parameters of loss function's distribution during training. Extensive experimental results on four benchmark datasets show that, without any tricks, our method outperforms existing hand-crafted loss functions in various computer vision tasks.
ISSN:2380-7504
DOI:10.1109/ICCV.2019.00850