Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
131 tokens/sec
GPT-4o
10 tokens/sec
Gemini 2.5 Pro Pro
47 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

Partial-label Learning with Mixed Closed-set and Open-set Out-of-candidate Examples (2307.00553v2)

Published 2 Jul 2023 in cs.LG and cs.CV

Abstract: Partial-label learning (PLL) relies on a key assumption that the true label of each training example must be in the candidate label set. This restrictive assumption may be violated in complex real-world scenarios, and thus the true label of some collected examples could be unexpectedly outside the assigned candidate label set. In this paper, we term the examples whose true label is outside the candidate label set OOC (out-of-candidate) examples, and pioneer a new PLL study to learn with OOC examples. We consider two types of OOC examples in reality, i.e., the closed-set/open-set OOC examples whose true label is inside/outside the known label space. To solve this new PLL problem, we first calculate the wooden cross-entropy loss from candidate and non-candidate labels respectively, and dynamically differentiate the two types of OOC examples based on specially designed criteria. Then, for closed-set OOC examples, we conduct reversed label disambiguation in the non-candidate label set; for open-set OOC examples, we leverage them for training by utilizing an effective regularization strategy that dynamically assigns random candidate labels from the candidate label set. In this way, the two types of OOC examples can be differentiated and further leveraged for model training. Extensive experiments demonstrate that our proposed method outperforms state-of-the-art PLL methods.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (52)
  1. Weakly supervised learning of instance segmentation with inter-pixel relations. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2209–2218.
  2. A. Bendale and T. Boult. 2016. Towards Open Set Deep Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.
  3. Learning and the unknown: Surveying steps toward open world recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. 9801–9807.
  4. Open-world semi-supervised learning. In Proceedings of the International Conference on Learning Representations.
  5. Learning open set network with discriminative reciprocal points. In Proceedings of the European Conference on Computer Vision. Springer, 507–522.
  6. Learning from partial labels. Journal of Machine Learning Research 12, 5 (2011), 1501–1536.
  7. Lei Feng and Bo An. 2019. Partial Label Learning by Semantic Difference Maximization.. In Proceedings of the International Joint Conference on Artificial Intelligence. 2294–2300.
  8. Provably consistent partial-label learning. In Advances in Neural Information Processing Systems.
  9. Recent Advances in Open Set Recognition: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence PP, 99 (2020), 1–1.
  10. Partial Label Learning via Label Influence Function. In Proceedings of the International Conference on Machine Learning. PMLR, 7665–7678.
  11. Co-teaching: Robust training of deep neural networks with extremely noisy labels. (2018).
  12. Eyke Hüllermeier and Jürgen Beringer. 2006. Learning from ambiguously labeled examples. Intell. Data Anal. 10, 5 (2006), 419–439.
  13. Multi-class open set recognition using probability of inclusion. In Proceeding of the European Conference on Computer Vision. Springer, 393–409.
  14. Nearest neighbors distance ratio open-set classifier. Machine Learning (2017).
  15. Shu Kong and Deva Ramanan. 2021. Opengan: Open-set recognition via open data generation. In Proceedings of the International Conference on Computer Vision. 813–822.
  16. Active partial label learning based on adaptive sample selection. International Journal of Machine Learning and Cybernetics 13, 6 (2022), 1603–1617.
  17. Towards safe weakly supervised learning. IEEE Transactions on Pattern Analysis and Machine Intelligence 43, 1 (2019), 334–346.
  18. Few-shot open-set recognition using meta-learning. In Proceedings of the Conference on Computer Vision and Pattern Recognition. 8798–8807.
  19. Pmal: Open set recognition via robust prototype mining. In Proceedings of the Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 36. 1872–1880.
  20. Progressive graph learning for open-set domain adaptation. In Proceedings of the International Conference on Machine Learning. PMLR, 6468–6478.
  21. Progressive identification of true labels for partial-label learning. In Proceedings of the International Conference on Machine Learning. 6500–6510.
  22. A self-paced regularization framework for partial-label learning. IEEE Transactions on Cybernetics (2020).
  23. GM-PLL: graph matching based partial label learning. IEEE Transactions on Knowledge and Data Engineering (2019).
  24. Poojan Oza and Vishal M Patel. 2019. C2ae: Class conditioned auto-encoder for open-set recognition. In Proceedings of the Conference on Computer Vision and Pattern Recognition. 2307–2316.
  25. Generative-discriminative feature representations for open-set recognition. In Proceedings of the Conference on Computer Vision and Pattern Recognition. 11814–11823.
  26. Tong Wei† Min-Ling Zhang Ruo-Jing Dong, Jun-Yi Hang. 2023. Can Label-Specific Features Help Partial-Label Learning?. In Proceedings of the Proceedings of the AAAI Conference on Artificial Intelligence.
  27. Evidentialmix: Learning with combined open-set and closed-set noisy labels. In Proceedings of the IEEE Winter Conference on Applications of Computer Vision. 3607–3615.
  28. Openmatch: Open-set semi-supervised learning with open-set consistency regularization. Advances in Neural Information Processing Systems 34 (2021), 25956–25967.
  29. Toward open set recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 35, 7 (2012), 1757–1772.
  30. Open-set learning under covariate shift. Machine Learning (2022), 1–17.
  31. Cai-Zhi Tang and Min-Ling Zhang. 2017. Confidence-rated discriminative partial label learning. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 31.
  32. Adaptive graph guided disambiguation for partial label learning. IEEE Transactions on Pattern Analysis and Machine Intelligence (2021).
  33. SoLar: Sinkhorn Label Refinery for Imbalanced Partial-Label Learning. Advances in Neural Information Processing Systems (2022).
  34. PiCO: Contrastive Label Disambiguation for Partial Label Learning. In Proceedings of the International Conference on Learning Representations.
  35. Wei Wang and Min-Ling Zhang. 2022. Partial Label Learning with Discrimination Augmentation. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1920–1928.
  36. Iterative learning with open-set noisy labels. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 8688–8696.
  37. Open-set label noise can improve robustness against inherent label noise. Advances in Neural Information Processing Systems 34 (2021), 7978–7992.
  38. Leveraged Weighted Loss for Partial Label Learning. In Proceedings of the International Conference on Machine Learning.
  39. Revisiting consistency regularization for deep partial label learning. In Proceedings of the International Conference on Machine Learning. PMLR, 24212–24225.
  40. NGC: A unified framework for learning with open-world noisy data. In Proceedings of the International Conference on Computer Vision. 62–71.
  41. Extended T: Learning with Mixed Closed-set and Open-set Noisy Labels. IEEE Transactions on Pattern Analysis and Machine Intelligence (2022).
  42. Partial label learning via label enhancement. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. 5557–5564.
  43. Instance-Dependent Partial Label Learning. Advances in Neural Information Processing Systems 34.
  44. Yan Yan and Yuhong Guo. 2020. Partial label learning with batch label correction. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 34. 6575–6582.
  45. Partial label learning with emerging new labels. Machine Learning (2022), 1–17.
  46. Understanding deep learning requires rethinking generalization. In Proceedings of the International Conference on Learning Representations.
  47. Exploiting Class Activation Value for Partial-Label Learning. In Proceedings of the International Conference on Learning Representations.
  48. Min-Ling Zhang and Fei Yu. 2015. Solving the partial label learning problem: An instance-based approach. In Proceedings of the International Joint Conference on Artificial Intelligence.
  49. Partial label learning via feature-aware disambiguation. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 1335–1344.
  50. Towards open-set object detection and discovery. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 3961–3970.
  51. Learning placeholders for open-set recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 4401–4410.
  52. Zhi-Hua Zhou. 2018. A brief introduction to weakly supervised learning. National Science Review 5, 1 (2018), 44–53.
Citations (1)
View on Semantic Scholar

Summary

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

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