反事实对比学习：通过因果图像合成实现鲁棒表征
Counterfactual contrastive learning: robust representations via causal image synthesis
AI总结：
针对对比预训练对增强流水线敏感的问题，提出利用反事实图像生成的CF-SimCLR方法，通过近似反事实推理构建正样本对，在医学影像数据集上提升了模型对采集偏移的鲁棒性与下游任务性能。
摘要：
对比预训练被广泛证实可提升下游任务性能与模型泛化能力，尤其是在标签有限的场景下。然而，它对增强流水线的选择十分敏感。正样本对应在保留语义信息的同时，去除领域特定信息。标准增强流水线通过预定义的光度变换模拟领域特定变化，但如果我们能模拟真实的领域变化呢？在本研究中，我们展示了如何利用反事实图像生成领域的最新进展来实现这一目标。我们提出CF-SimCLR，这是一种反事实对比学习方法，利用近似反事实推理来构建正样本对。在胸部X光和乳腺钼靶的5个数据集上的综合评估表明，CF-SimCLR大幅提升了对采集偏移的鲁棒性，在分布内和分布外数据上均取得了更优的下游性能，对训练中代表性不足的领域效果尤为显著。
英文摘要：
Contrastive pretraining is well-known to improve downstream task performance and model generalisation, especially in limited label settings. However, it is sensitive to the choice of augmentation pipeline. Positive pairs should preserve semantic information while destroying domain-specific information. Standard augmentation pipelines emulate domain-specific changes with pre-defined photometric transformations, but what if we could simulate realistic domain changes instead? In this work, we show how to utilise recent progress in counterfactual image generation to this effect. We propose CF-SimCLR, a counterfactual contrastive learning approach which leverages approximate counterfactual inference for positive pair creation. Comprehensive evaluation across five datasets, on chest radiography and mammography, demonstrates that CF-SimCLR substantially improves robustness to acquisition shift with higher downstream performance on both in- and out-of-distribution data, particularly for domains which are under-represented during training.
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。
