用混合数据模拟异常样本，同时解决类别不平衡和分布外检测问题。
RICASSO: Reinforced Imbalance Learning with Class-Aware Self-Supervised Outliers Exposure
- 用数据混合生成伪异常数据，替代真实异常数据进行训练。
- 在iNaturalist2018上实现AUROC提升27%、假阳性率降低61%。
- 仅使用正常数据即可完成异常暴露，适合真实场景部署。
现实场景中，深度学习模型常面临类别不平衡（长尾）与分布外（OOD）数据的双重挑战。现有联合方法依赖真实OOD数据，导致性能妥协。本文发现，数据混合这一强大增强技术可生成兼具分布内（ID）与分布外（OOD）特征的伪OOD数据。因此，通过使用混合数据而非真实OOD数据，可统一解决长尾识别与OOD检测问题。我们提出统一框架RICASSO（Reinforced Imbalance Learning with Class-Aware Self-Supervised Outliers Exposure），其中“自监督”指仅使用ID数据进行异常暴露。RICASSO包含三项核心策略：基于范数-奇异性对偶的异常暴露，利用混合数据通过单一损失函数同步实现ID数据重平衡与异常暴露；基于模糊性的逻辑值调整，利用ID数据的模糊性自适应校准输出；对比边界-中心学习，结合虚拟边界学习与双熵中心学习，利用混合数据实现更优特征分离与聚类，并通过表示一致性学习提升鲁棒性。大量实验表明，RICASSO在长尾识别上达到当前最优性能，且相比基线在iNaturalist2018上实现AUROC提升27%、假阳性率降低61%。在iNaturalist2018上，其表现甚至优于使用真实OOD数据的方法。代码即将公开。
原文摘要 · Abstract (English)
In real-world scenarios, deep learning models often face challenges from both imbalanced (long-tailed) and out-of-distribution (OOD) data. However, existing joint methods rely on real OOD data, which leads to unnecessary trade-offs. In contrast, our research shows that data mixing, a potent augmentation technique for long-tailed recognition, can generate pseudo-OOD data that exhibit the features of both in-distribution (ID) data and OOD data. Therefore, by using mixed data instead of real OOD data, we can address long-tailed recognition and OOD detection holistically. We propose a unified framework called Reinforced Imbalance Learning with Class-Aware Self-Supervised Outliers Exposure (RICASSO), where "self-supervised" denotes that we only use ID data for outlier exposure. RICASSO includes three main strategies: Norm-Odd-Duality-Based Outlier Exposure: Uses mixed data as pseudo-OOD data, enabling simultaneous ID data rebalancing and outlier exposure through a single loss function. Ambiguity-Aware Logits Adjustment: Utilizes the ambiguity of ID data to adaptively recalibrate logits. Contrastive Boundary-Center Learning: Combines Virtual Boundary Learning and Dual-Entropy Center Learning to use mixed data for better feature separation and clustering, with Representation Consistency Learning for robustness. Extensive experiments demonstrate that RICASSO achieves state-of-the-art performance in long-tailed recognition and significantly improves OOD detection compared to our baseline (27% improvement in AUROC and 61% reduction in FPR on the iNaturalist2018 dataset). On iNaturalist2018, we even outperforms methods using real OOD data. The code will be made public soon.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。