面向真实测试时自适应的批归一化解构研究
Unraveling Batch Normalization for Realistic Test-Time Adaptation
AI总结：
针对真实小批量下测试时自适应因目标估计不准性能下降的问题，本文解构批归一化发现批次类别多样性不足是核心原因，提出TEMA工具与逐层校正策略，无需训练调参即可提升域偏移鲁棒性，在多基准上达SOTA。
摘要：
尽管现有测试时自适应方法通过调整批归一化缩小域差异展现出一定效果，但在真实小批量场景下，由于目标估计不准确，其有效性会大幅下降。以往研究仅通过引入源域统计量来缓解这一问题，目标估计不准确的根本问题仍未解决，测试时的固有域偏移也未得到处理。本文深入探究小批量导致的性能退化问题，通过解构批归一化发现，不精确的目标域统计量很大程度上源于批次内类别多样性大幅降低。基于这一洞见，我们提出一种简洁的工具——测试时指数移动平均（Test-time Exponential Moving Average, TEMA），以弥补训练批次与测试批次之间的类别多样性差距。值得注意的是，TEMA 自适应地将常规方法的作用范围从当前批次拓展出去，融入多样化的类别信息，进而提升目标估计的准确性。在此基础上，我们进一步设计了一种新颖的逐层校正策略，持续提升测试时性能。本文方法具有独特优势：既无需训练也无需调参，是一种真正零负担的解决方案。它显著增强了模型对域偏移的鲁棒性，在不同批量大小的各类真实场景中均保持良好适应性，在多个主流基准上取得了当前最优性能。代码可在 https://github.com/kiwi12138/RealisticTTA 获取。
英文摘要：
While recent test-time adaptations exhibit efficacy by adjusting batch normalization to narrow domain disparities, their effectiveness diminishes with realistic mini-batches due to inaccurate target estimation. As previous attempts merely introduce source statistics to mitigate this issue, the fundamental problem of inaccurate target estimation still persists, leaving the intrinsic test-time domain shifts unresolved. This paper delves into the problem of mini-batch degradation. By unraveling batch normalization, we discover that the inexact target statistics largely stem from the substantially reduced class diversity in batch. Drawing upon this insight, we introduce a straightforward tool, Test-time Exponential Moving Average (TEMA), to bridge the class diversity gap between training and testing batches. Importantly, our TEMA adaptively extends the scope of typical methods beyond the current batch to incorporate a diverse set of class information, which in turn boosts an accurate target estimation. Built upon this foundation, we further design a novel layer-wise rectification strategy to consistently promote test-time performance. Our proposed method enjoys a unique advantage as it requires neither training nor tuning parameters, offering a truly hassle-free solution. It significantly enhances model robustness against shifted domains and maintains resilience in diverse real-world scenarios with various batch sizes, achieving state-of-the-art performance on several major benchmarks. Code is available at \url{https://github.com/kiwi12138/RealisticTTA}.
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。
