Abstract
Current contrastive learning methods use random transformations sampled from a large list of transformations, with fixed hyper-parameters, to learn invariance from an unannotated database. Following previous works that introduce a small amount of supervision, we propose a framework to find optimal transformations for contrastive learning using a differentiable transformation network. Our method increases performances at low annotated data regime both in supervision accuracy and in convergence speed. In contrast to previous work, no generative model is needed for transformation optimization. Transformed images keep relevant information to solve the supervised task, here classification. Experiments were performed on 34000 2D slices of brain Magnetic Resonance Images and 11200 chest X-ray images. On both datasets, with 10% of labeled data, our model achieves better performances than a fully supervised model with 100% labels.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Chaitanya, K., Erdil, E., Karani, N., Konukoglu, E.: Contrastive learning of global and local features for medical image segmentation with limited annotations. In: NeurIPS, vol. 33, pp. 12546–12558 (2020)
Chen, T., Kornblith, S., Norouzi, M., Hinton, G.: A simple framework for contrastive learning of visual representations. In: 37th International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 119, pp. 1597–1607. PMLR (2020)
Cohen, J.P., Luck, M., Honari, S.: Distribution matching losses can hallucinate features in medical image translation. In: MICCAI (2018)
Cubuk, E.D., Zoph, B., Mane, D., Vasudevan, V., Le, Q.V.: AutoAugment: learning augmentation strategies from data. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 113–123 (2019)
Dufumier, B., et al.: Contrastive learning with continuous proxy meta-data for 3D MRI classification. In: MICCAI (2021)
Khosla, P., et al.: Supervised contrastive learning. In: Advances in Neural Information Processing Systems, vol. 33, pp. 18661–18673 (2020)
Kingma, D.P., Dhariwal, P.: Glow: generative flow with invertible 1x1 convolutions. In: NeurIPS, vol. 31 (2018)
Li, Y., Hu, G., Wang, Y., Hospedales, T., Robertson, N.M., Yang, Y.: Differentiable automatic data augmentation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12367, pp. 580–595. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58542-6_35
Liu, A., Huang, Z., Huang, Z., Wang, N.: Direct differentiable augmentation search. In: 2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2021)
Menze, B.H., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imaging 34(10), 1993–2024 (2015)
Patrick, M., et al.: On compositions of transformations in contrastive self-supervised learning. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp. 9577–9587 (2021)
Perakis, A., Gorji, A., Jain, S., Chaitanya, K., Rizza, S., Konukoglu, E.: Contrastive learning of single-cell phenotypic representations for treatment classification. In: MLMI - MICCAI, pp. 565–575 (2021)
Tang, Y., et al.: Automated abnormality classification of chest radiographs using deep convolutional neural networks. NPJ Digit. Med. 3, 1–8 (2020)
Tian, Y., Sun, C., Poole, B., Krishnan, D., Schmid, C., Isola, P.: What makes for good views for contrastive learning? In: NeurIPS (2020)
Xiao, T., Wang, X., Efros, A.A., Darrell, T.: What should not be contrastive in contrastive learning. In: International Conference on Learning Representations (2021)
Yang, S., Das, D., Chang, S., Yun, S., Porikli, F.M.: Distribution estimation to automate transformation policies for self-supervision. In: Advances in Neural Information Processing Systems (2021)
Zhao, S., Liu, Z., Lin, J., Zhu, J.Y., Han, S.: Differentiable augmentation for data-efficient GAN training. In: Advances in Neural Information Processing Systems, vol. 33, pp. 7559–7570 (2020)
Zhao, X., et al.: Contrastive learning for label efficient semantic segmentation. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp. 10623–10633 (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Ruppli, C., Gori, P., Ardon, R., Bloch, I. (2022). Optimizing Transformations for Contrastive Learning in a Differentiable Framework. In: Zamzmi, G., Antani, S., Bagci, U., Linguraru, M.G., Rajaraman, S., Xue, Z. (eds) Medical Image Learning with Limited and Noisy Data. MILLanD 2022. Lecture Notes in Computer Science, vol 13559. Springer, Cham. https://doi.org/10.1007/978-3-031-16760-7_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-16760-7_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16759-1
Online ISBN: 978-3-031-16760-7
eBook Packages: Computer ScienceComputer Science (R0)