Academic Journal
Adaptive sample repulsion against class-specific counterfactuals for explainable imbalanced classification.
| Τίτλος: | Adaptive sample repulsion against class-specific counterfactuals for explainable imbalanced classification. |
|---|---|
| Συγγραφείς: | Hao Y; School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunications, Beijing, 100876, China. Electronic address: haoyu_bupt@bupt.edu.cn., Gao X; School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunications, Beijing, 100876, China. Electronic address: xlhhh74@bupt.edu.cn., Diao X; Metrology Research Institute, China Electric Power Research Institute Company Limited, Beijing, 100192, China. Electronic address: diaoxp@epri.sgcc.com.cn., Li Y; Metrology Research Institute, China Electric Power Research Institute Company Limited, Beijing, 100192, China. Electronic address: liyuan3@epri.sgcc.com.cn., Lin Y; Metrology Research Institute, China Electric Power Research Institute Company Limited, Beijing, 100192, China. Electronic address: linyukun@epri.sgcc.com.cn., Chen T; Metrology Research Institute, China Electric Power Research Institute Company Limited, Beijing, 100192, China. Electronic address: chentianyang@epri.sgcc.com.cn., Li Q; School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunications, Beijing, 100876, China. Electronic address: 2782709791@bupt.edu.cn., Lu J; School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunications, Beijing, 100876, China. Electronic address: lujiawen@bupt.edu.cn. |
| Πηγή: | Neural networks : the official journal of the International Neural Network Society [Neural Netw] 2026 Jul; Vol. 199, pp. 108652. Date of Electronic Publication: 2026 Jan 30. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Pergamon Press Country of Publication: United States NLM ID: 8805018 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1879-2782 (Electronic) Linking ISSN: 08936080 NLM ISO Abbreviation: Neural Netw Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: New York : Pergamon Press, [c1988- |
| Ιατρικοί όροι (MeSH): | Classification Algorithms*, Data Mining/methods ; Reinforcement Machine Learning |
| Περίληψη: | Enhancing model classification capability for samples within overlapping regions in complex feature spaces remains a key challenge in imbalanced classification research. Existing mainstream methods at the data-level and algorithm-level primarily rely on original sample distribution information to reduce overlap impact, without deeply modeling the causal relationship between features and labels. Furthermore, these approaches often overlook instance-level explanations that could guide deep discriminative information mining for samples of different classes in overlapping regions, thus the improvement on classification performance and model credibility may be constrained. This paper proposes an explainable imbalanced classification framework with adaptive sample repulsion against class-specific counterfactuals (CSCF-SR), forming a closed-loop between explanation generation and classification decisions by dynamically regulating the feature-space distribution through generated counterfactual samples. Two core phases are jointly optimized. (1) Counterfactual searching: a class-specific dual-actor architecture based on reinforcement learning decouples perturbation policy learning for majority and minority classes. A multi-step dynamic perturbation mechanism is designed to control counterfactual search behavior more precisely and smoothly, effectively generating reliable counterfactual samples. (2) Adaptive sample repulsion against counterfactuals: exploiting the inter-class discriminative information in displacement vectors between counterfactual and original samples, each original sample is adaptively perturbed along the direction opposite to its counterfactual. This fine-grained regulation gradually displaces samples from the overlapping region and clarifies class boundaries. Experiments on 50 imbalanced datasets demonstrate that CSCF-SR has a performance advantage over 27 typical imbalanced classification methods on both F1-score and G-mean, with more pronounced improvements on 25 datasets with severe class overlap. (Copyright © 2026 Elsevier Ltd. All rights reserved.) |
| Competing Interests: | Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. |
| Contributed Indexing: | Keywords: Counterfactual search; Explainable machine learning; Imbalanced classification; Inter-class overlap; Sample distribution control |
| Entry Date(s): | Date Created: 20260204 Date Completed: 20260706 Latest Revision: 20260707 |
| Update Code: | 20260707 |
| DOI: | 10.1016/j.neunet.2026.108652 |
| PMID: | 41638095 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1879-2782 |
|---|---|
| DOI: | 10.1016/j.neunet.2026.108652 |