SoDeep: a Sorting Deep net to learn ranking loss surrogates
Autor: | Matthieu Cord, Patrick Pérez, Louis Chevallier, Martin Engilberge |
---|---|
Přispěvatelé: | Engilberge, Martin, Technicolor R & I [Cesson Sévigné], Technicolor, Machine Learning and Information Access (MLIA), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), InterDigital Communications, Valeo.ai, VALEO |
Jazyk: | angličtina |
Rok vydání: | 2019 |
Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Computer science Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition [INFO.INFO-NE] Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] Machine Learning (stat.ML) 02 engineering and technology [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] 010501 environmental sciences [INFO] Computer Science [cs] Machine learning computer.software_genre 01 natural sciences Synthetic data Computer Science - Information Retrieval Machine Learning (cs.LG) Ranking (information retrieval) [INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] Statistics - Machine Learning 0202 electrical engineering electronic engineering information engineering [INFO]Computer Science [cs] Proxy (statistics) 0105 earth and related environmental sciences Flexibility (engineering) Contextual image classification business.industry Deep learning Sorting [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG] Metric (mathematics) 020201 artificial intelligence & image processing Artificial intelligence business computer Information Retrieval (cs.IR) |
Zdroj: | CVPR CVPR 2019-2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPR 2019-2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2019, Long Beach, United States |
Popis: | Several tasks in machine learning are evaluated using non-differentiable metrics such as mean average precision or Spearman correlation. However, their non-differentiability prevents from using them as objective functions in a learning framework. Surrogate and relaxation methods exist but tend to be specific to a given metric. In the present work, we introduce a new method to learn approximations of such non-differentiable objective functions. Our approach is based on a deep architecture that approximates the sorting of arbitrary sets of scores. It is trained virtually for free using synthetic data. This sorting deep (SoDeep) net can then be combined in a plug-and-play manner with existing deep architectures. We demonstrate the interest of our approach in three different tasks that require ranking: Cross-modal text-image retrieval, multi-label image classification and visual memorability ranking. Our approach yields very competitive results on these three tasks, which validates the merit and the flexibility of SoDeep as a proxy for sorting operation in ranking-based losses. Comment: Accepted to CVPR 2019 |
Databáze: | OpenAIRE |
Externí odkaz: |