Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 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

Rethinking Multi-domain Generalization with A General Learning Objective (2402.18853v2)

Published 29 Feb 2024 in cs.LG, cs.AI, and cs.CV

Abstract: Multi-domain generalization (mDG) is universally aimed to minimize the discrepancy between training and testing distributions to enhance marginal-to-label distribution mapping. However, existing mDG literature lacks a general learning objective paradigm and often imposes constraints on static target marginal distributions. In this paper, we propose to leverage a $Y$-mapping to relax the constraint. We rethink the learning objective for mDG and design a new \textbf{general learning objective} to interpret and analyze most existing mDG wisdom. This general objective is bifurcated into two synergistic amis: learning domain-independent conditional features and maximizing a posterior. Explorations also extend to two effective regularization terms that incorporate prior information and suppress invalid causality, alleviating the issues that come with relaxed constraints. We theoretically contribute an upper bound for the domain alignment of domain-independent conditional features, disclosing that many previous mDG endeavors actually \textbf{optimize partially the objective} and thus lead to limited performance. As such, our study distills a general learning objective into four practical components, providing a general, robust, and flexible mechanism to handle complex domain shifts. Extensive empirical results indicate that the proposed objective with $Y$-mapping leads to substantially better mDG performance in various downstream tasks, including regression, segmentation, and classification.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. Invariant risk minimization. arXiv preprint arXiv:1907.02893, 2019.
  2. Recognition in terra incognita. In Proceedings of the European conference on computer vision (ECCV), pages 456–473, 2018.
  3. Generalizing from several related classification tasks to a new unlabeled sample. Advances in neural information processing systems, 24, 2011.
  4. Domain generalization by marginal transfer learning. The Journal of Machine Learning Research, 22(1):46–100, 2021.
  5. Exploiting domain-specific features to enhance domain generalization. Advances in Neural Information Processing Systems, 34:21189–21201, 2021.
  6. Swad: Domain generalization by seeking flat minima. Advances in Neural Information Processing Systems, 34:22405–22418, 2021.
  7. Learning to balance specificity and invariance for in and out of domain generalization. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX 16, pages 301–318. Springer, 2020.
  8. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European conference on computer vision (ECCV), pages 801–818, 2018.
  9. Cooperative distribution alignment via jsd upper bound. Advances in Neural Information Processing Systems, 35:21101–21112, 2022.
  10. Robustnet: Improving domain generalization in urban-scene segmentation via instance selective whitening. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11580–11590, 2021.
  11. The cityscapes dataset for semantic urban scene understanding. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3213–3223, 2016.
  12. Unbiased metric learning: On the utilization of multiple datasets and web images for softening bias. In Proceedings of the IEEE International Conference on Computer Vision, pages 1657–1664, 2013.
  13. Domain-adversarial training of neural networks. The journal of machine learning research, 17(1):2096–2030, 2016.
  14. In search of lost domain generalization. arXiv preprint arXiv:2007.01434, 2020.
  15. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  16. Domain generalization via multidomain discriminant analysis. In Uncertainty in Artificial Intelligence, pages 292–302. PMLR, 2020.
  17. Self-challenging improves cross-domain generalization. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part II 16, pages 124–140. Springer, 2020.
  18. Domain generalization by mutual-information regularization with pre-trained models. European Conference on Computer Vision (ECCV), 2022.
  19. Style neophile: Constantly seeking novel styles for domain generalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7130–7140, 2022.
  20. Steven M Kay. Fundamentals of statistical signal processing: estimation theory. Prentice-Hall, Inc., 1993.
  21. Selfreg: Self-supervised contrastive regularization for domain generalization. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9619–9628, 2021.
  22. Out-of-distribution generalization via risk extrapolation (rex). In International Conference on Machine Learning, pages 5815–5826. PMLR, 2021.
  23. Wildnet: Learning domain generalized semantic segmentation from the wild. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9936–9946, 2022.
  24. Deeper, broader and artier domain generalization. In Proceedings of the IEEE international conference on computer vision, pages 5542–5550, 2017.
  25. Learning to generalize: Meta-learning for domain generalization. In Proceedings of the AAAI conference on artificial intelligence, 2018a.
  26. Domain generalization with adversarial feature learning. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5400–5409, 2018b.
  27. Domain generalization via conditional invariant representations. In Proceedings of the AAAI conference on artificial intelligence, 2018c.
  28. Deep domain generalization via conditional invariant adversarial networks. In Proceedings of the European conference on computer vision (ECCV), pages 624–639, 2018d.
  29. Simple: Specialized model-sample matching for domain generalization. In The Eleventh International Conference on Learning Representations, 2022.
  30. Jianhua Lin. Divergence measures based on the shannon entropy. IEEE Transactions on Information theory, 37(1):145–151, 1991.
  31. Va-depthnet: A variational approach to single image depth prediction. arXiv preprint arXiv:2302.06556, 2023.
  32. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 10012–10022, 2021.
  33. Reducing domain gap by reducing style bias. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8690–8699, 2021.
  34. The mapillary vistas dataset for semantic understanding of street scenes. In Proceedings of the IEEE international conference on computer vision, pages 4990–4999, 2017.
  35. Two at once: Enhancing learning and generalization capacities via ibn-net. In Proceedings of the European Conference on Computer Vision (ECCV), pages 464–479, 2018.
  36. Semantic-aware domain generalized segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2594–2605, 2022.
  37. Moment matching for multi-source domain adaptation. In Proceedings of the IEEE/CVF international conference on computer vision, pages 1406–1415, 2019.
  38. Empirical risk minimization with relative entropy regularization: Optimality and sensitivity analysis. In 2022 IEEE International Symposium on Information Theory (ISIT), pages 684–689. IEEE, 2022.
  39. Diverse weight averaging for out-of-distribution generalization. Advances in Neural Information Processing Systems, 35:10821–10836, 2022.
  40. Playing for data: Ground truth from computer games. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part II 14, pages 102–118. Springer, 2016.
  41. The synthia dataset: A large collection of synthetic images for semantic segmentation of urban scenes. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3234–3243, 2016.
  42. Distributionally robust neural networks for group shifts: On the importance of regularization for worst-case generalization. arXiv preprint arXiv:1911.08731, 2019.
  43. Gradient matching for domain generalization. arXiv preprint arXiv:2104.09937, 2021.
  44. Hidetoshi Shimodaira. Improving predictive inference under covariate shift by weighting the log-likelihood function. Journal of statistical planning and inference, 90(2):227–244, 2000.
  45. Indoor segmentation and support inference from rgbd images. In Computer Vision–ECCV 2012: 12th European Conference on Computer Vision, Florence, Italy, October 7-13, 2012, Proceedings, Part V 12, pages 746–760. Springer, 2012.
  46. Revisiting weakly supervised pre-training of visual perception models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 804–814, 2022.
  47. Deep coral: Correlation alignment for deep domain adaptation. In Computer Vision–ECCV 2016 Workshops: Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III 14, pages 443–450. Springer, 2016.
  48. Vladimir N. Vapnik. Statistical Learning Theory. Wiley-Interscience, 1998.
  49. Deep hashing network for unsupervised domain adaptation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5018–5027, 2017.
  50. Generalizing to unseen domains: A survey on domain generalization. IEEE Transactions on Knowledge and Data Engineering, 2022.
  51. Bdd100k: A diverse driving dataset for heterogeneous multitask learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2636–2645, 2020.
  52. Domain adaptation under target and conditional shift. In International conference on machine learning, pages 819–827. PMLR, 2013.
  53. Multi-source domain adaptation: A causal view. In Proceedings of the AAAI Conference on Artificial Intelligence, 2015.
  54. Adaptive risk minimization: Learning to adapt to domain shift. Advances in Neural Information Processing Systems, 34:23664–23678, 2021.
  55. Domain generalization with mixstyle. arXiv preprint arXiv:2104.02008, 2021.
  56. Domain generalization: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022.
Citations (7)
View on Semantic Scholar

Summary

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

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