Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
125 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Continuous Unsupervised Domain Adaptation Using Stabilized Representations and Experience Replay (2402.00580v1)

Published 31 Jan 2024 in cs.LG and cs.CV

Abstract: We introduce an algorithm for tackling the problem of unsupervised domain adaptation (UDA) in continual learning (CL) scenarios. The primary objective is to maintain model generalization under domain shift when new domains arrive continually through updating a base model when only unlabeled data is accessible in subsequent tasks. While there are many existing UDA algorithms, they typically require access to both the source and target domain datasets simultaneously. Conversely, existing CL approaches can handle tasks that all have labeled data. Our solution is based on stabilizing the learned internal distribution to enhances the model generalization on new domains. The internal distribution is modeled by network responses in hidden layer. We model this internal distribution using a Gaussian mixture model (GMM ) and update the model by matching the internally learned distribution of new domains to the estimated GMM. Additionally, we leverage experience replay to overcome the problem of catastrophic forgetting, where the model loses previously acquired knowledge when learning new tasks. We offer theoretical analysis to explain why our algorithm would work. We also offer extensive comparative and analytic experiments to demonstrate that our method is effective. We perform experiments on four benchmark datasets to demonstrate that our approach is effective.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (93)
  1. Memory aware synapses: Learning what (not) to forget. In Proceedings of the European conference on computer vision (ECCV), pages 139–154, 2018.
  2. Concrete problems in ai safety. arXiv preprint arXiv:1606.06565, 2016.
  3. Wasserstein GAN. arXiv preprint arXiv:1701.07875, 2017.
  4. Deepjdot: Deep joint distribution optimal transport for unsupervised domain adaptation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 447–463, 2018.
  5. Adapting to continuously shifting domains. 2018.
  6. N. Bonnotte. Unidimensional and evolution methods for optimal transportation. PhD thesis, Paris 11, 2013.
  7. Task-attentive transformer architecture for continual learning of vision-and-language tasks using knowledge distillation. arXiv preprint arXiv:2303.14423, 2023.
  8. Joint domain alignment and discriminative feature learning for unsupervised deep domain adaptation. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 3296–3303, 2019.
  9. Lifelong machine learning. Synthesis Lectures on Artificial Intelligence and Machine Learning, 12(3):1–207, 2018.
  10. Optimal transport for domain adaptation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(9):1853–1865, 2016.
  11. Domain generalization via model-agnostic learning of semantic features. Advances in neural information processing systems, 32, 2019.
  12. Robert M French. Catastrophic forgetting in connectionist networks. Trends in cognitive sciences, 3(4):128–135, 1999.
  13. Domain-adversarial training of neural networks. The Journal of Machine Learning Research, 17(1):2096–2030, 2016.
  14. Unsupervised domain adaptation by backpropagation. In Proceedings of International Conference on Machine Learning, pages 1180–1189, 2015.
  15. Unsupervised domain adaptation via deep conditional adaptation network. Pattern Recognition, 134:109088, 2023.
  16. Deep reconstruction-classification networks for unsupervised domain adaptation. In European Conference on Computer Vision, pages 597–613. Springer, 2016.
  17. An empirical investigation of catastrophic forgetting in gradient-based neural networks. arXiv preprint arXiv:1312.6211, 2013.
  18. A survey on adversarial domain adaptation. Neural Processing Letters, 55(3):2429–2469, 2023.
  19. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 770–778, 2016.
  20. S. Helgason. The radon transform on rn. In Integral Geometry and Radon Transforms, pages 1–62. Springer, 2011.
  21. CyCADA: Cycle-consistent adversarial domain adaptation. In International Conference on Machine Learning, pages 1989–1998, 2018.
  22. Domain-invariant representation learning using an unsupervised domain adversarial adaptation deep neural network. Neurocomputing, 355:209–220, 2019.
  23. Unsupervised domain adaptation for training event-based networks using contrastive learning and uncorrelated conditioning. arXiv preprint arXiv:2303.12424, 2023.
  24. Learn continually, generalize rapidly: Lifelong knowledge accumulation for few-shot learning. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 714–729, 2021.
  25. Deep generative dual memory network for continual learning. arXiv preprint arXiv:1710.10368, 2017.
  26. Contrastive adaptation network for unsupervised domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 4893–4902, 2019.
  27. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of sciences, 114(13):3521–3526, 2017.
  28. Sliced wasserstein kernels for probability distributions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 5258–5267, 2016.
  29. Universal source-free domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4544–4553, 2020.
  30. Deep learning. nature, 521(7553):436–444, 2015.
  31. Sliced wasserstein discrepancy for unsupervised domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 10285–10295, 2019.
  32. Continual learning for domain adaptation in chest x-ray classification. In Medical Imaging with Deep Learning, pages 413–423. PMLR, 2020.
  33. Enhanced transport distance for unsupervised domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13936–13944, 2020.
  34. Adaptive batch normalization for practical domain adaptation. Pattern Recognition, 80:109–117, 2018.
  35. Do we really need to access the source data? source hypothesis transfer for unsupervised domain adaptation. In International Conference on Machine Learning, pages 6028–6039. PMLR, 2020.
  36. Source data-absent unsupervised domain adaptation through hypothesis transfer and labeling transfer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(11):8602–8617, 2021.
  37. Coupled generative adversarial networks. In Advances in neural information processing systems, pages 469–477, 2016.
  38. Yu Han Liu. Feature extraction and image recognition with convolutional neural networks. In Journal of Physics: Conference Series, volume 1087, page 062032. IOP Publishing, 2018.
  39. Learning transferable features with deep adaptation networks. In Proceedings of International Conference on Machine Learning, pages 97–105, 2015.
  40. Conditional adversarial domain adaptation. In Advances in Neural Information Processing Systems, pages 1640–1650, 2018.
  41. Deep transfer learning with joint adaptation networks. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages 2208–2217. JMLR. org, 2017.
  42. Gradient episodic memory for continual learning. In Advances in neural information processing systems, pages 6467–6476, 2017.
  43. Deep multi-modality adversarial networks for unsupervised domain adaptation. IEEE Transactions on Multimedia, 21(9):2419–2431, 2019.
  44. Analysis of multidimensional xor classification problem with evolutionary feedforward neural networks. International Journal on Artificial Intelligence Tools, 16(01):111–120, 2007.
  45. UMAP: Uniform manifold approximation and projection. Journal of Open Source Software, 3(29):861, 2018.
  46. History repeats: Overcoming catastrophic forgetting for event-centric temporal knowledge graph completion. arXiv preprint arXiv:2305.18675, 2023.
  47. Minimal-entropy correlation alignment for unsupervised deep domain adaptation. In ICLR, 2018.
  48. Properties of artificial networks evolved to contend with natural spectra. Proceedings of the National Academy of Sciences, 111(Supplement 3):10868–10872, 2014.
  49. R. Neal. Slice sampling. Annals of statistics, pages 705–741, 2003.
  50. Unsupervised domain adaptation of object detectors: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.
  51. Transferrable prototypical networks for unsupervised domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2239–2247, 2019.
  52. Multi-adversarial domain adaptation. In Proceedings Thirty-Second AAAI Conference on Artificial Intelligence, pages 3934–3941, 2018.
  53. Domain2vec: Domain embedding for unsupervised domain adaptation. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VI 16, pages 756–774. Springer, 2020.
  54. Don’t worry about the weather: Unsupervised condition-dependent domain adaptation. In 2019 IEEE Intelligent Transportation Systems Conference (ITSC), pages 33–40. IEEE, 2019.
  55. The challenge of data annotation in deep learning—a case study on whole plant corn silage. Sensors, 22(4):1596, 2022.
  56. icarl: Incremental classifier and representation learning. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pages 2001–2010, 2017.
  57. A. Redko, I.and Habrard and M. Sebban. Theoretical analysis of domain adaptation with optimal transport. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pages 737–753. Springer, 2017.
  58. Learning to learn without forgetting by maximizing transfer and minimizing interference. In International Conference on Learning Representations, 2018.
  59. Anthony Robins. Catastrophic forgetting, rehearsal and pseudorehearsal. Connection Science, 7(2):123–146, 1995.
  60. Mohammad Rostami. Lifelong domain adaptation via consolidated internal distribution. Advances in neural information processing systems, 34:11172–11183, 2021.
  61. Domain adaptation for sentiment analysis using robust internal representations. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 11484–11498, 2023.
  62. Cognitively inspired learning of incremental drifting concepts. In 2023 International Joint Conference on Artificial Intelligence, 2023.
  63. Overcoming concept shift in domain-aware settings through consolidated internal distributions. In Proceedings of the AAAI conference on artificial intelligence, volume 1, 2023.
  64. A crowdsourcing triage algorithm for geopolitical event forecasting. In Proceedings of the 12th ACM Conference on Recommender Systems, pages 377–381. ACM, 2018.
  65. Deep transfer learning for few-shot sar image classification. Remote Sensing, 11(11):1374, 2019.
  66. Sar image classification using few-shot cross-domain transfer learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pages 0–0, 2019.
  67. Generative continual concept learning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 5545–5552, 2020.
  68. Complementary learning for overcoming catastrophic forgetting using experience replay. In Proceedings of the 28th International Joint Conference on Artificial Intelligence, pages 3339–3345. AAAI Press, 2019.
  69. Progressive neural networks. arXiv preprint arXiv:1606.04671, 2016.
  70. Generate to adapt: Aligning domains using generative adversarial networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 8503–8512, 2018.
  71. Prioritized experience replay. arXiv preprint arXiv:1511.05952, 2015.
  72. Large-scale optimal transport and mapping estimation. In International Conference on Learning Representations, 2018.
  73. Understanding machine learning: From theory to algorithms. Cambridge university press, 2014.
  74. Cnn features off-the-shelf: an astounding baseline for recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition workshops, pages 806–813, 2014.
  75. Wasserstein distance guided representation learning for domain adaptation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32, 2018.
  76. Learning across domains and devices: Style-driven source-free domain adaptation in clustered federated learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 444–454, 2023.
  77. Continual learning with deep generative replay. In Advances in Neural Information Processing Systems, pages 2990–2999, 2017.
  78. Prototypical networks for few-shot learning. Advances in neural information processing systems, 30, 2017.
  79. Climb: A continual learning benchmark for vision-and-language tasks. Advances in Neural Information Processing Systems, 35:29440–29453, 2022.
  80. Unsupervised model adaptation for continual semantic segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 2593–2601, 2021.
  81. Domain adaptation for the segmentation of confidential medical images. In British Machine Vision Conference, 2023.
  82. Return of frustratingly easy domain adaptation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 30, 2016.
  83. Deep coral: Correlation alignment for deep domain adaptation. In European conference on computer vision, pages 443–450. Springer, 2016.
  84. Adversarial discriminative domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 7167–7176, 2017.
  85. Online verification of deep neural networks under domain or weight shift. arXiv preprint arXiv:2106.12732, 2021.
  86. Ace: Adapting to changing environments for semantic segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 2121–2130, 2019.
  87. Incremental adversarial domain adaptation for continually changing environments. In 2018 IEEE International conference on robotics and automation (ICRA), pages 4489–4495. IEEE, 2018.
  88. Adaptive progressive continual learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(10):6715–6728, 2021.
  89. Generalized source-free domain adaptation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8978–8987, 2021.
  90. Central moment discrepancy (cmd) for domain-invariant representation learning. In ICLR, 2017.
  91. Continual learning through synaptic intelligence. In International Conference on Machine Learning, pages 3987–3995. PMLR, 2017.
  92. Source-free unsupervised domain adaptation: Current research and future directions. Neurocomputing, page 126921, 2023.
  93. Domain-symmetric networks for adversarial domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 5031–5040, 2019.
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