Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Adaptive Multi-head Contrastive Learning (2310.05615v3)

Published 9 Oct 2023 in cs.CV, cs.AI, and cs.LG

Abstract: In contrastive learning, two views of an original image, generated by different augmentations, are considered a positive pair, and their similarity is required to be high. Similarly, two views of distinct images form a negative pair, with encouraged low similarity. Typically, a single similarity measure, provided by a lone projection head, evaluates positive and negative sample pairs. However, due to diverse augmentation strategies and varying intra-sample similarity, views from the same image may not always be similar. Additionally, owing to inter-sample similarity, views from different images may be more akin than those from the same image. Consequently, enforcing high similarity for positive pairs and low similarity for negative pairs may be unattainable, and in some cases, such enforcement could detrimentally impact performance. To address this challenge, we propose using multiple projection heads, each producing a distinct set of features. Our pre-training loss function emerges from a solution to the maximum likelihood estimation over head-wise posterior distributions of positive samples given observations. This loss incorporates the similarity measure over positive and negative pairs, each re-weighted by an individual adaptive temperature, regulated to prevent ill solutions. Our approach, Adaptive Multi-Head Contrastive Learning (AMCL), can be applied to and experimentally enhances several popular contrastive learning methods such as SimCLR, MoCo, and Barlow Twins. The improvement remains consistent across various backbones and linear probing epochs, and becomes more significant when employing multiple augmentation methods.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (51)
  1. A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information fusion, 76:243–297, 2021.
  2. Neural machine translation by jointly learning to align and translate. In Yoshua Bengio and Yann LeCun (eds.), 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings, 2015. URL http://arxiv.org/abs/1409.0473.
  3. A cookbook of self-supervised learning. arXiv preprint arXiv:2304.12210, 2023.
  4. BEit: BERT pre-training of image transformers. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=p-BhZSz59o4.
  5. VICReg: Variance-invariance-covariance regularization for self-supervised learning. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=xm6YD62D1Ub.
  6. Hyperopt: a python library for model selection and hyperparameter optimization. Computational Science & Discovery, 8(1):014008, 2015. URL http://stacks.iop.org/1749-4699/8/i=1/a=014008.
  7. Unsupervised learning of visual features by contrasting cluster assignments. Advances in neural information processing systems, 33:9912–9924, 2020.
  8. Emerging properties in self-supervised vision transformers. In Proceedings of the International Conference on Computer Vision (ICCV), 2021.
  9. A simple framework for contrastive learning of visual representations. In International conference on machine learning, pp. 1597–1607. PMLR, 2020.
  10. Exploring simple siamese representation learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  15750–15758, 2021.
  11. Attention-based models for speech recognition. Advances in neural information processing systems, 28, 2015.
  12. An analysis of single-layer networks in unsupervised feature learning. In Geoffrey Gordon, David Dunson, and Miroslav Dudík (eds.), Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, volume 15 of Proceedings of Machine Learning Research, pp.  215–223, Fort Lauderdale, FL, USA, 11–13 Apr 2011. PMLR. URL https://proceedings.mlr.press/v15/coates11a.html.
  13. Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp.  248–255, 2009. doi: 10.1109/CVPR.2009.5206848.
  14. Self-contrastive learning with hard negative sampling for self-supervised point cloud learning. In Proceedings of the 29th ACM International Conference on Multimedia, pp.  3133–3142, 2021.
  15. With a little help from my friends: Nearest-neighbor contrastive learning of visual representations. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  9588–9597, 2021.
  16. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In international conference on machine learning, pp. 1050–1059. PMLR, 2016.
  17. A survey of uncertainty in deep neural networks. Artificial Intelligence Review, pp.  1–77, 2023.
  18. Bayesian neural networks: An introduction and survey. Case Studies in Applied Bayesian Data Science: CIRM Jean-Morlet Chair, Fall 2018, pp.  45–87, 2020.
  19. Bootstrap your own latent-a new approach to self-supervised learning. Advances in neural information processing systems, 33:21271–21284, 2020.
  20. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  770–778, 2016.
  21. Momentum contrast for unsupervised visual representation learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  9729–9738, 2020.
  22. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  16000–16009, 2022.
  23. Deep metric learning using triplet network. In Similarity-Based Pattern Recognition: Third International Workshop, SIMBAD 2015, Copenhagen, Denmark, October 12-14, 2015. Proceedings 3, pp.  84–92. Springer, 2015.
  24. Harold Hotelling. Relations between two sets of variates. Biometrika, 28(3/4):321–377, 1936. ISSN 00063444. URL http://www.jstor.org/stable/2333955.
  25. Towards the generalization of contrastive self-supervised learning. In The Eleventh International Conference on Learning Representations, 2022a.
  26. Contrastive masked autoencoders are stronger vision learners. arXiv preprint arXiv:2207.13532, 2022b.
  27. Robust statistics. Wiley New York, 1981.
  28. Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods. Mach. Learn., 110(3):457–506, 2021. doi: 10.1007/s10994-021-05946-3.
  29. Layer grafted pre-training: Bridging contrastive learning and masked image modeling for label-efficient representations. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=jwdqNwyREyh.
  30. What uncertainties do we need in bayesian deep learning for computer vision? In I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (eds.), Advances in Neural Information Processing Systems, volume 30. Curran Associates, Inc., 2017.
  31. Supervised contrastive learning. Advances in neural information processing systems, 33:18661–18673, 2020.
  32. Mean shift for self-supervised learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp.  10326–10335, October 2021.
  33. Alex Krizhevsky. Learning multiple layers of features from tiny images. 2009. URL https://api.semanticscholar.org/CorpusID:18268744.
  34. Temperature schedules for self-supervised contrastive methods on long-tail data. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=ejHUr4nfHhD.
  35. Ya Le and Xuan S. Yang. Tiny imagenet visual recognition challenge. 2015. URL https://api.semanticscholar.org/CorpusID:16664790.
  36. Hermann G. Matthies. Quantifying uncertainty: Modern computational representation of probability and applications. In Adnan Ibrahimbegovic and Ivica Kozar (eds.), Extreme Man-Made and Natural Hazards in Dynamics of Structures, pp.  105–135, Dordrecht, 2007. Springer Netherlands. ISBN 978-1-4020-5656-7.
  37. A simple, efficient and scalable contrastive masked autoencoder for learning visual representations. arXiv preprint arXiv:2210.16870, 2022.
  38. Multi-similarity contrastive learning. arXiv preprint arXiv:2307.02712, 2023.
  39. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023.
  40. What do self-supervised vision transformers learn? In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=azCKuYyS74.
  41. Sean Tao. Deep neural network ensembles. In Machine Learning, Optimization, and Data Science: 5th International Conference, LOD 2019, Siena, Italy, September 10–13, 2019, Proceedings 5, pp.  1–12. Springer, 2019.
  42. Understanding the behaviour of contrastive loss. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  2495–2504, 2021.
  43. Uncertainty-dtw for time series and sequences. In European Conference on Computer Vision, pp.  176–195. Springer, 2022.
  44. Understanding contrastive representation learning through alignment and uniformity on the hypersphere. In International Conference on Machine Learning, pp. 9929–9939. PMLR, 2020.
  45. Contrastive learning rivals masked image modeling in fine-tuning via feature distillation. arXiv preprint arXiv:2205.14141, 2022.
  46. Unsupervised feature learning via non-parametric instance discrimination. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  3733–3742, 2018.
  47. Simmim: A simple framework for masked image modeling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  9653–9663, 2022.
  48. Barlow twins: Self-supervised learning via redundancy reduction. In International Conference on Machine Learning, pp. 12310–12320. PMLR, 2021.
  49. Temperature as uncertainty in contrastive learning. arXiv preprint arXiv:2110.04403, 2021.
  50. Generalized cross entropy loss for training deep neural networks with noisy labels. Advances in neural information processing systems, 31, 2018.
  51. ibot: Image bert pre-training with online tokenizer. arXiv preprint arXiv:2111.07832, 2021.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (4)
  1. Lei Wang (975 papers)
  2. Piotr Koniusz (84 papers)
  3. Tom Gedeon (72 papers)
  4. Liang Zheng (181 papers)
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