几何约束异常合成
Geometrically Constrained Outlier Synthesis
AI总结：
提出几何约束异常合成（GCOS）方法，通过在隐藏特征空间中生成受几何约束的虚拟异常样本，结合对比正则化，提升图像分类模型对分布外样本的鲁棒性。
摘要：
用于图像分类的深度神经网络通常对分布外（OOD）样本表现出过度自信。为了解决这个问题，我们引入了几何约束异常合成（GCOS），这是一种训练时正则化框架，旨在提高推理时的OOD鲁棒性。GCOS通过生成隐藏特征空间中尊重分布内（ID）数据学习到的流形结构的虚拟异常，解决了先前合成方法的局限性。合成分两个阶段进行：（i）从训练特征中提取的主方差子空间识别出几何信息引导的离流形方向；（ii）由校准集中非一致性得分的经验分位数定义的一个类共形壳，自适应地控制合成幅度以产生边界样本。该壳确保生成的异常既不是微不足道可检测的，也不是与分布内数据无法区分的，从而促进更平滑地学习鲁棒特征。这与对比正则化目标相结合，在选定的得分空间（如马氏距离或基于能量的）中促进ID和OOD样本的可分离性。实验表明，在近OOD基准测试（定义为异常与分布内数据共享相同语义域的任务）上，使用标准基于能量的推理时，GCOS优于最先进的方法。作为探索性扩展，该框架自然地过渡到共形OOD推理，将不确定性得分转化为统计上有效的p值，并启用具有形式误差保证的阈值，为更可预测和可靠的OOD检测提供了途径。
英文摘要：
Deep neural networks for image classification often exhibit overconfidence on out-of-distribution (OOD) samples. To address this, we introduce Geometrically Constrained Outlier Synthesis (GCOS), a training-time regularization framework aimed at improving OOD robustness during inference. GCOS addresses a limitation of prior synthesis methods by generating virtual outliers in the hidden feature space that respect the learned manifold structure of in-distribution (ID) data. The synthesis proceeds in two stages: (i) a dominant-variance subspace extracted from the training features identifies geometrically informed, off-manifold directions; (ii) a conformally-inspired shell, defined by the empirical quantiles of a nonconformity score from a calibration set, adaptively controls the synthesis magnitude to produce boundary samples. The shell ensures that generated outliers are neither trivially detectable nor indistinguishable from in-distribution data, facilitating smoother learning of robust features. This is combined with a contrastive regularization objective that promotes separability of ID and OOD samples in a chosen score space, such as Mahalanobis or energy-based. Experiments demonstrate that GCOS outperforms state-of-the-art methods using standard energy-based inference on near-OOD benchmarks, defined as tasks where outliers share the same semantic domain as in-distribution data. As an exploratory extension, the framework naturally transitions to conformal OOD inference, which translates uncertainty scores into statistically valid p-values and enables thresholds with formal error guarantees, providing a pathway toward more predictable and reliable OOD detection.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
