提出新方法提升集合表示在推理时元素损坏下的鲁棒性。
Distributionally Robust Set Representation Learning Under Inference-Time Element Corruption
- 用可微分的单纯形权重优化模拟最坏情况下的元素扰动。
- 在四个任务中均有效提升抗干扰能力且保持高性能。
- 适合部署在存在数据缺失或异常的现实场景中使用。
标准集合表示学习方法在精心整理的数据上表现优异，但常忽略推理时元素受损的问题。这指部署模型遇到元素级退化（如异常值或缺失成分）时，可能扭曲集合表示并降低性能。我们提出SW-DRSO，一种专为集合设计的分布鲁棒优化框架。不只最小化训练数据上的损失，还优化一个可计算的、对可能推理时变化的最坏期望损失的近似。引入基于质心的对抗者，通过可微分的单纯形权重优化，在训练时逼近不可计算的受损集合搜索。在四个任务上的大量实验表明，SW-DRSO能有效增强对扰动的鲁棒性，同时维持高整体性能。
原文摘要 · Abstract (English)
Standard Set Representation Learning methods typically excel on curated data but often overlook the challenge of inference-time element corruption. This refers to scenarios where deployed models encounter element-level degradations, such as outliers or missing components, that may distort set representation and degrade performance. We propose SW-DRSO, a distributionally robust optimization framework tailored for sets. Rather than minimizing loss solely on observed training data, SW-DRSO optimizes a tractable surrogate of the worst-case expected loss over a family of plausible inference-time variations. We introduce a barycentric adversary that approximates the intractable search over corrupted sets by a differentiable training-time optimization over simplex weights. Extensive experiments across four tasks demonstrate that SW-DRSO effectively enhances robustness against corruption while maintaining high overall performance.
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。