Partial-Label Regression (2306.08968v1)
Abstract: Partial-label learning is a popular weakly supervised learning setting that allows each training example to be annotated with a set of candidate labels. Previous studies on partial-label learning only focused on the classification setting where candidate labels are all discrete, which cannot handle continuous labels with real values. In this paper, we provide the first attempt to investigate partial-label regression, where each training example is annotated with a set of real-valued candidate labels. To solve this problem, we first propose a simple baseline method that takes the average loss incurred by candidate labels as the predictive loss. The drawback of this method lies in that the loss incurred by the true label may be overwhelmed by other false labels. To overcome this drawback, we propose an identification method that takes the least loss incurred by candidate labels as the predictive loss. We further improve it by proposing a progressive identification method to differentiate candidate labels using progressively updated weights for incurred losses. We prove that the latter two methods are model-consistent and provide convergence analyses. Our proposed methods are theoretically grounded and can be compatible with any models, optimizers, and losses. Experiments validate the effectiveness of our proposed methods.
- Support vector machines for multiple-instance learning. In NeurIPS.
- Semi-Supervised Learning. MIT Press.
- Learning from ambiguously labeled face images. TPAMI, 40(7): 1653–1667.
- Chowdhary, K. 2020. Natural language processing. Fundamentals of Artificial Intelligence, 603–649.
- Learning from partial labels. Journal of Machine Learning Research, 12(5): 1501–1536.
- Learning classifiers from only positive and unlabeled data. In KDD, 213–220.
- Partial Label Learning with Self-Guided Retraining. In AAAI, 3542–3549.
- Provably consistent partial-label learning. In NeurIPS, 10948–10960.
- Robust loss functions under label noise for deep neural networks. In AAAI.
- A regularization approach for instance-based superset label learning. IEEE Transactions on Cybernetics, 48(3): 967–978.
- Discriminative metric learning for partial label learning. TNNLS.
- Top-k partial label machine. TNNLS.
- Partial Label Learning via Label Influence Function. In ICML, 7665–7678.
- Co-teaching: Robust training of deep neural networks with extremely noisy labels. In NeurIPS, 8527–8537.
- Complementary-label learning for arbitrary losses and models. In ICML, 2971–2980.
- Learning with multiple labels. In NeurIPS, 921–928.
- Adam: A method for stochastic optimization. In ICLR.
- Positive-unlabeled learning with non-negative risk estimator. In NeurIPS, 1674–1684.
- Classification with noisy labels by importance reweighting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(3): 447–461.
- Metric learning for multi-output tasks. TPAMI, 41(2): 408–422.
- Learning from candidate labeling sets. In NeurIPS, 1504–1512.
- Progressive Identification of True Labels for Partial-Label Learning. In ICML.
- Decoupling” when to update” from” how to update”. In NeurIPS, volume 30.
- A framework for multiple-instance learning. In NeurIPS.
- Foundations of Machine Learning. MIT Press.
- Classification with partial labels. In KDD, 551–559.
- Theoretical comparisons of positive-unlabeled learning against positive-negative learning. In NeurIPS, volume 29.
- Making deep neural networks robust to label noise: A loss correction approach. In CVPR, 1944–1952.
- Fixmatch: Simplifying semi-supervised learning with consistency and confidence. In NeurIPS, 596–608.
- Szeliski, R. 2010. Computer vision: algorithms and applications. Springer Science & Business Media.
- Vapnik, V. N. 1999. An overview of statistical learning theory. IEEE Transactions on Neural Networks, 10(5): 988–999.
- Adaptive Graph Guided Disambiguation for Partial Label Learning. In KDD, 83–91.
- PiCO: Contrastive Label Disambiguation for Partial Label Learning. In ICLR.
- Combating noisy labels by agreement: A joint training method with co-regularization. In CVPR, 13726–13735.
- Leveraged weighted loss for partial label learning. In ICML, 11091–11100.
- Maximum margin partial label learning. In ACML, 96–111.
- Learning by associating ambiguously labeled images. In CVPR, 708–715.
- Exploiting Class Activation Value for Partial-Label Learning. In ICLR.
- Solving the Partial Label Learning Problem: An Instance-Based Approach. In IJCAI, 4048–4054.
- Partial label learning via feature-aware disambiguation. In KDD, 1335–1344.