论文标题
通过有条件不变的正常化程序破坏相关转移
Breaking Correlation Shift via Conditional Invariant Regularizer
论文作者
论文摘要
最近，对分布式（OOD）数据的概括引起了人们的关注。相关转移是由与类标签相关的虚假属性引起的，因为它们之间的相关性可能在培训和测试数据中有所不同。对于这样一个问题，我们表明，鉴于类标签，有条件地独立于虚假属性的模型是可推广的。基于此，提出了控制OOD泛化误差的度量有条件伪变异（CSV），以衡量这种条件独立性。为了改善OOD的概括，我们将训练过程正常使用，该培训过程与拟议的CSV相关。在温和的假设下，我们的训练目标可以作为非convex-concave mini-max问题提出。提出了具有可证明的收敛速率的算法来解决该问题。广泛的经验结果验证了我们算法在改善OOD概括方面的功效。
Recently, generalization on out-of-distribution (OOD) data with correlation shift has attracted great attentions. The correlation shift is caused by the spurious attributes that correlate to the class label, as the correlation between them may vary in training and test data. For such a problem, we show that given the class label, the models that are conditionally independent of spurious attributes are OOD generalizable. Based on this, a metric Conditional Spurious Variation (CSV) which controls the OOD generalization error, is proposed to measure such conditional independence. To improve the OOD generalization, we regularize the training process with the proposed CSV. Under mild assumptions, our training objective can be formulated as a nonconvex-concave mini-max problem. An algorithm with a provable convergence rate is proposed to solve the problem. Extensive empirical results verify our algorithm's efficacy in improving OOD generalization.
