Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Mean Field Theory in Deep Metric Learning (2306.15368v1)

Published 27 Jun 2023 in cs.LG and cond-mat.stat-mech

Abstract: In this paper, we explore the application of mean field theory, a technique from statistical physics, to deep metric learning and address the high training complexity commonly associated with conventional metric learning loss functions. By adapting mean field theory for deep metric learning, we develop an approach to design classification-based loss functions from pair-based ones, which can be considered complementary to the proxy-based approach. Applying the mean field theory to two pair-based loss functions, we derive two new loss functions, MeanFieldContrastive and MeanFieldClassWiseMultiSimilarity losses, with reduced training complexity. We extensively evaluate these derived loss functions on three image-retrieval datasets and demonstrate that our loss functions outperform baseline methods in two out of the three datasets.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (38)
  1. Learning fine-grained image similarity with deep ranking. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1386–1393, 2014.
  2. Facenet: A unified embedding for face recognition and clustering. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 815–823, 2015.
  3. In defense of the triplet loss for person re-identification. arXiv preprint arXiv:1703.07737, 2017.
  4. Philip W Anderson. More is different: broken symmetry and the nature of the hierarchical structure of science. Science, 177(4047):393–396, 1972.
  5. Pierre Weiss. L’hypothèse du champ moléculaire et la propriété ferromagnétique. J. Phys. Theor. Appl., 6(1):661–690, 1907.
  6. No fuss distance metric learning using proxies. In Proceedings of the IEEE international conference on computer vision, pages 360–368, 2017.
  7. A metric learning reality check. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXV 16, pages 681–699. Springer, 2020a.
  8. Proxy anchor loss for deep metric learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3238–3247, 2020.
  9. Dimensionality reduction by learning an invariant mapping. In 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’06), volume 2, pages 1735–1742. IEEE, 2006.
  10. Distance metric learning for large margin nearest neighbor classification. Journal of machine learning research, 10(2), 2009.
  11. Kihyuk Sohn. Improved deep metric learning with multi-class n-pair loss objective. Advances in neural information processing systems, 29, 2016.
  12. Deep metric learning via lifted structured feature embedding. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4004–4012, 2016.
  13. Ranked list loss for deep metric learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5207–5216, 2019a.
  14. Multi-similarity loss with general pair weighting for deep metric learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5022–5030, 2019b.
  15. Embedding deep metric for person re-identification: A study against large variations. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14, pages 732–748. Springer, 2016.
  16. Sampling matters in deep embedding learning. In Proceedings of the IEEE international conference on computer vision, pages 2840–2848, 2017.
  17. Hard-aware deeply cascaded embedding. In Proceedings of the IEEE international conference on computer vision, pages 814–823, 2017.
  18. Smart mining for deep metric learning. In Proceedings of the IEEE International Conference on Computer Vision, pages 2821–2829, 2017.
  19. Normface: L2 hypersphere embedding for face verification. In Proceedings of the 25th ACM international conference on Multimedia, pages 1041–1049, 2017.
  20. Classification is a strong baseline for deep metric learning. arXiv preprint arXiv:1811.12649, 2018.
  21. Sphereface: Deep hypersphere embedding for face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 212–220, 2017.
  22. Arcface: Additive angular margin loss for deep face recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4690–4699, 2019.
  23. Additive margin softmax for face verification. IEEE Signal Processing Letters, 25(7):926–930, 2018a.
  24. Cosface: Large margin cosine loss for deep face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5265–5274, 2018b.
  25. Fewer is more: A deep graph metric learning perspective using fewer proxies. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 17792–17803, 2020.
  26. Hierarchical proxy-based loss for deep metric learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 1859–1868, 2022.
  27. Softtriple loss: Deep metric learning without triplet sampling. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6450–6458, 2019.
  28. Proxynca++: Revisiting and revitalizing proxy neighborhood component analysis. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIV 16, pages 448–464. Springer, 2020.
  29. Informative sample-aware proxy for deep metric learning. In Proceedings of the 4th ACM International Conference on Multimedia in Asia, pages 1–11, 2022.
  30. Elements of phase transitions and critical phenomena. Oup Oxford, 2010.
  31. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, 32, 2019.
  32. Pytorch metric learning. ArXiv, abs/2008.09164, 2020b.
  33. The caltech-ucsd birds-200-2011 dataset. 2011.
  34. 3d object representations for fine-grained categorization. In Proceedings of the IEEE international conference on computer vision workshops, pages 554–561, 2013.
  35. Deepfashion: Powering robust clothes recognition and retrieval with rich annotations. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1096–1104, 2016.
  36. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International conference on machine learning, pages 448–456. pmlr, 2015.
  37. Imagenet large scale visual recognition challenge. International journal of computer vision, 115:211–252, 2015.
  38. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now