用于领域自适应与领域泛化的保序一致性正则化
Order-preserving Consistency Regularization for Domain Adaptation and Generalization
AI总结：
本文提出保序一致性正则化（OCR），通过保持预测概率的顺序不变来增强模型对任务无关变换的鲁棒性，从而降低对领域特定属性的敏感度，并在五个跨领域任务上验证了其优势。
摘要：
如果深度学习模型对领域特定属性（例如光照、背景、相机角度等）过度敏感，就会在跨领域挑战中失败。为了缓解这一问题，通常采用数据增强与一致性正则化相结合的方法，使模型对领域特定属性不那么敏感。一致性正则化强制模型对同一图像的两个视图输出相同的表示或预测。然而，这些约束要么过于严格，要么对分类概率不保序。在这项工作中，我们提出了用于跨领域任务的保序一致性正则化（OCR）。预测的保序特性使模型对与任务无关的变换具有鲁棒性。因此，模型对领域特定属性的敏感度降低。综合实验表明，我们的方法在五个不同的跨领域任务上取得了明显优势。
英文摘要：
Deep learning models fail on cross-domain challenges if the model is oversensitive to domain-specific attributes, e.g., lightning, background, camera angle, etc. To alleviate this problem, data augmentation coupled with consistency regularization are commonly adopted to make the model less sensitive to domain-specific attributes. Consistency regularization enforces the model to output the same representation or prediction for two views of one image. These constraints, however, are either too strict or not order-preserving for the classification probabilities. In this work, we propose the Order-preserving Consistency Regularization (OCR) for cross-domain tasks. The order-preserving property for the prediction makes the model robust to task-irrelevant transformations. As a result, the model becomes less sensitive to the domain-specific attributes. The comprehensive experiments show that our method achieves clear advantages on five different cross-domain tasks.
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。
