Meta-Learning for Domain Generalization Under Distribution Shift: Methods, Benchmarks, and Open Challenges

References

[1] Bai, H., Sun, R., Hong, L., Zhou, F., Ye, N., Ye, H.J., Chan, S.H.G., Li, Z.: DecAug: Out-of-distribution generalization via decomposed feature representation and se-mantic augmentation. In: Proceedings of the 35th AAAI Conference on Artificial Intelligence (2021). https://doi.org/10.1609/aaai.v35i8.16829

[2] Balaji, Y., Sankaranarayanan, S., Chellappa, R.: MetaReg: Towards domain gener-alization using meta-regularization. In: Advances in Neural Information Processing Systems. vol. 31 (2018)

[3] Chen, J., Gao, Z., Wu, X., Luo, J.: Meta-causal learning for sin-gle domain generalization. arXiv preprint arXiv:2304.03709 (2023). https://doi.org/10.48550/arXiv.2304.03709

[4] Chi, Z., Gu, L., Zhong, T., Liu, H., Yu, Y., Plataniotis, K.N., Wang, Y.: Adapt-ing to distribution shift by visual domain prompt generation. arXiv preprint arXiv:2405.02797 (2024). https://doi.org/10.48550/arXiv.2405.02797

[5] Dong, Y., Gong, T., Chen, H., Song, S., Zhang, W., Li, C.: How does distribu-tion matching help domain generalization: An information-theoretic analysis. arXiv preprint arXiv:2406.09745 (2024). https://doi.org/10.48550/arXiv.2406.09745

[6] Du, Y., Xu, J., Xiong, H., Qiu, Q., Zhen, X., Snoek, C.G.M., Shao, L.: Learning to learn with variational information bottleneck for domain generalization. arXiv preprint arXiv:2007.07645 (2020). https://doi.org/10.48550/arXiv.2007.07645

[7] Enomoto, S.: Pseudo multi-source domain generalization: Bridging the gap between single and multi-source domain generalization. arXiv preprint arXiv:2505.23173 (2025). https://doi.org/10.48550/arXiv.2505.23173

[8] Faryna, K., van der Laak, J., Litjens, G.: Automatic data augmen-tation to improve generalization of deep learning in h&e-stained histopathology. Computers in Biology and Medicine 170, 108018 (2024). https://doi.org/10.1016/j.compbiomed.2024.108018

[9] Gholamzadeh Khoee, A., Yu, Y., Feldt, R.: Domain generalization through meta-learning: A survey. Artificial Intelligence Review (2024). https://doi.org/10.1007/s10462-024-10922-z

[10] Gulrajani, I., López-Paz, D.: In search of lost domain generalization. arXiv preprint arXiv:2007.01434 (2020). https://doi.org/10.48550/arXiv.2007.01434

[11] Hospedales, T.M., Antoniou, A., Micaelli, P., Storkey, A.: Meta-learning in neural networks: A survey. IEEE Transactions on Pattern Analysis and Machine Intelli-gence 44(9), 5149–5169 (2021). https://doi.org/10.1109/tpami.2021.3079209

[12] Koh, P.W., Sagawa, S., Marklund, H., Xie, S.M., Zhang, M., Balsubramani, A., Hu, W., Yasunaga, M., Phillips, R.L., Gao, I., Lee, T., David, E., Stavness, I., Guo, W., Earnshaw, B., Haque, I., Beery, S.M., Leskovec, J., Kundaje, A., Pierson, E., Levine, S., Finn, C., Liang, P.: WILDS: A bench-mark of in-the-wild distribution shifts. arXiv preprint arXiv:2012.07421 (2020). https://doi.org/10.48550/arXiv.2012.07421

[13] Li, C., Liu, Y., Li, M., Li, X., Song, Y., Yu, Y.: Domain generalization on med-ical imaging classification using episodic training with task augmentation. arXiv preprint arXiv:2106.06908 (2021). https://doi.org/10.48550/arXiv.2106.06908

[14] Li, D., Yang, Y., Song, Y.Z., Hospedales, T.M.: Learning to generalize: Meta-learning for domain generalization. In: Proceedings of the 32nd AAAI Conference on Artificial Intelligence (2018). https://doi.org/10.1609/aaai.v32i1.11596

[15] Liang, J., He, R., Tan, T.: A comprehensive survey on test-time adap-tation under distribution shifts. arXiv preprint arXiv:2303.15361 (2023). https://doi.org/10.48550/arXiv.2303.15361

[16] Liao, Y., et al.: Episodic training and feature orthogonality-driven domain gener-alization for rotating machinery fault diagnosis under unseen working conditions. Machines 13(7), 563 (2025). https://doi.org/10.3390/machines13070563

[17] Lv, F., Liang, J., Li, S., Zang, B., Liu, C.H., Wang, Z., Liu, D.: Causality inspired representation learning for domain generalization. arXiv preprint arXiv:2203.14237 (2022). https://doi.org/10.48550/arXiv.2203.14237

[18] Nguyen, C.Q., Kreatsoulas, C., Branson, K.M.: Meta-learning GNN initializations for low-resource molecular property prediction. arXiv preprint arXiv:2003.05996 (2020). https://doi.org/10.48550/arXiv.2003.05996

[19] Shen, Z., Yu, H., Cui, P., Liu, J., Zhang, X., Zhou, L., Liu, F.: Meta adaptive task sampling for few-domain generalization. arXiv preprint arXiv:2305.15644 (2023). https://doi.org/10.48550/arXiv.2305.15644

[20] Shi, Y., Seely, J., Torr, P.H.S., Siddharth, N., Hannun, A., Usunier, N., Synnaeve, G.: Gradient matching for domain generalization. arXiv preprint arXiv:2104.09937 (2021). https://doi.org/10.48550/arXiv.2104.09937

[21] Tian, Q., Zhao, C., Shao, M., Wang, W., Lin, Y., Li, D.: MLDGG: Meta-learning for domain generalization on graphs. arXiv preprint arXiv:2411.12913 (2024). https://doi.org/10.48550/arXiv.2411.12913

[22] Wang, H., Deng, Z.: Cross-domain few-shot classification via adversarial task aug-mentation. In: Proceedings of the 30th International Joint Conference on Artificial Intelligence. pp. 1–7 (2021). https://doi.org/10.24963/ijcai.2021/149

[23] Wang, P., Zhang, Z., Lei, Z., Zhang, L.: Sharpness-aware gradient matching for domain generalization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2023). https://doi.org/10.1109/cvpr52729.2023.00367

[24] Wang, X., Peng, D., Hu, P., Sang, J.: Generalizable decision boundaries: Dualistic meta-learning for open set domain generalization. arXiv preprint arXiv:2308.09391 (2023). https://doi.org/10.48550/arXiv.2308.09391

[25] Wei, Y., Han, Y.: Multi-source collaborative gradient discrepancy minimization for federated domain generalization. In: Proceedings of the 38th AAAI Conference on Artificial Intelligence (2024). https://doi.org/10.1609/aaai.v38i14.29510

[26] Wong, G., et al.: Weighted risk invariance: Domain generalization un-der invariant feature shift. arXiv preprint arXiv:2407.18428 (2024). https://doi.org/10.48550/arXiv.2407.18428

[27] Wu, Y., Chen, Z., Wang, W., Liu, Y.: Test-time domain adaptation by learning domain-aware batch normalization. In: Proceedings of the 38th AAAI Conference on Artificial Intelligence (2024). https://doi.org/10.1609/aaai.v38i14.29527

[28] Xiao, Z., Shen, X., Zhen, X., van den Hengel, A., Shao, L., Snoek, C.G.M.: Learning to generalize across domains on single test samples. arXiv preprint arXiv:2202.08045 (2022). https://doi.org/10.48550/arXiv.2202.08045

[29] Xin, S., Wang, Y., Li, J., Guo, Y., Ding, P., Li, W.: On the connection between invariant learning and adversarial training for out-of-distribution generalization. In: Proceedings of the 37th AAAI Conference on Artificial Intelligence (2023). https://doi.org/10.1609/aaai.v37i9.26250

[30] Yang, S., Wang, Y., Joao Ribeiro, D., Bhatt, R., Lim, S.N., Torr, P.H.S.: Open domain generalization with domain-augmented meta-learning. arXiv preprint arXiv:2104.03620 (2021). https://doi.org/10.48550/arXiv.2104.03620

[31] Zheng, G., Huai, M., Zhang, A.: AdvST: Revisiting data augmentations for single domain generalization. In: Proceedings of the 38th AAAI Conference on Artificial Intelligence (2024). https://doi.org/10.1609/aaai.v38i19.30184

@article{1719292,
      author = {Aryanil Roy, Mainak Ghatak, Sananda Chatterjee, Kaushik Banerjee},
      title = {Meta-Learning for Domain Generalization Under Distribution Shift: Methods, Benchmarks, and Open Challenges},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {12},
      pages = {3230-3238},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1719292.pdf},
      abstract = {Domain generalization (DG) explores how a model trained on a fixed set of source domains can perform reliably on unseen tar-get domains. Meta-learning addresses this by training models to explic-itly learn to generalize, yet curated benchmarks have repeatedly shown that well-tuned evidence-based risk minimization (ERM) is a formidable baseline. This survey examines that tension with a focus on what ac-tually generalizes and why. We contribute three things. First, a bivari-ate taxonomy that cross references shift type (covariate, conditional, in-variant covariate, category, compound) against meta intervention level (data/augmentation, representation/gradient, optimization/parameter, prompt/foundation). Second, a structured comparison of benchmark fam-ilies: DomainBed, WILDS, single source, and open set settings that shows how benchmark choice, and not just method design, drives published conclusions. Third, a critical analysis of conditions under which meta-learning gains over ERM are real versus aphemeral, updated for the 2023-2026 period when foundation models, causal approaches, and meta prompting have substantially changed the landscape. We conclude with actionable open challenges and directions where meta-learning retains a genuine edge.},
      keywords = {Domain Generalization, Meta-Learning, Distribution Shift, Domainbed, WILDS, Benchmark Comparison, Transfer Learning},
      month = {June},
      doi = {https://doi.org/10.64388/IREV9I12-1719292}
  }