面向恶劣条件下域自适应语义分割的启发式自步学习
Heuristic Self-Paced Learning for Domain Adaptive Semantic Segmentation under Adverse Conditions
把课程学习建模为序贯决策问题，用高维状态编码器+策略梯度自动调度类别学习顺序，可直接迁移到任何需要按难度/信息量安排训练样本的视觉任务（如域自适应检测、分割）。
原文依据 we cast curriculum learning as a sequential decision problem and propose an autonomous class scheduler
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论语义类别的学习顺序显著影响面向语义分割的无监督域自适应，在恶劣天气条件下尤为如此。现有课程大多依赖手工设计的启发式规则（例如固定的不确定性度量）并遵循静态调度，无法适应模型不断演化的高维训练动态，从而导致类别偏差。受强化学习启发，我们将课程学习建模为序贯决策问题，并提出一种自主类别调度器。该调度器由两个部分组成：(i) 一个高维状态编码器，将模型的训练状态映射到潜空间，并提炼出指示训练进展的关键特征；(ii) 一个类别公平的策略梯度目标，确保各类别获得均衡提升。结合混合源-目标监督，学习到的类别排序引导网络在每个阶段聚焦于最具信息量的类别，从而实现更具自适应性和动态性的学习。值得注意的是，我们的方法在三个广泛使用的基准（例如 ACDC、Dark Zurich 和 Nighttime Driving）上取得了最优性能，并在合成到真实的语义分割中展现出泛化能力。
英文原文摘要 1240 字符
The learning order of semantic classes significantly impacts unsupervised domain adaptation for semantic segmentation, especially under adverse weather conditions. Most existing curricula rely on handcrafted heuristics (e.g., fixed uncertainty metrics) and follow a static schedule, which fails to adapt to a model's evolving, high-dimensional training dynamics, leading to category bias. Inspired by Reinforcement Learning, we cast curriculum learning as a sequential decision problem and propose an autonomous class scheduler. This scheduler consists of two components: (i) a high-dimensional state encoder that maps the model's training status into a latent space and distills key features indicative of progress, and (ii) a category-fair policy-gradient objective that ensures balanced improvement across classes. Coupled with mixed source-target supervision, the learned class rankings direct the network's focus to the most informative classes at each stage, enabling more adaptive and dynamic learning. It is worth noting that our method achieves state-of-the-art performance on three widely used benchmarks (e.g., ACDC, Dark Zurich, and Nighttime Driving) and shows generalization ability in synthetic-to-real semantic segmentation.