Shift multiaccuracy 为分布偏移学习提供更优误差基准
Multiaccuracy as an Error Benchmark for Learning with Distribution Shift
搞理论或做过域适应的朋友值得看这篇:它修了 TDS 框架里误差被常数倍放大的老毛病,借了公平性领域的 multiaccuracy 来定义新基准,还能继承 agnostic learning 的下界,域适应方向可以拿来用。
arXiv 论文 2611.19701 针对 testable learning with distribution shift(TDS)场景提出了新的误差基准 shift multiaccuracy。现有 agnostic TDS 的标准保证即使在训练与测试分布完全一致时,仍允许误差相比最优值有常数倍的放大,因此无法直接退化回经典的 agnostic learning 保证。
论文借鉴算法公平性领域的 multiaccuracy 概念定义了 shift multiaccuracy:在无分布偏移时,该 TDS 框架与 multiaccuracy 等价。利用已有工作中从 agnostic learning 到 multiaccurate prediction 的非平凡归约,作者得到首个能通过保训练边缘分布的高效归约正式推广 agnostic learning 的 TDS 框架,保留了现有算法上界并继承 agnostic learning 的计算复杂性下界。论文还建立了该模型与 identity testing 的直接联系,首次给出 agnostic TDS 学习基于 Rademacher complexity 的统计刻画。作者指出 shift multiaccuracy 在无监督域适应中有独立价值,可带来更紧的 OOD 泛化保证,且其定义不涉及不可观测的测试标签,因而可以直接最小化。
Multiaccuracy as an Error Benchmark for Learning with Distribution Shift
Agnostic learning has a well-developed theory of computational and statistical complexity. For learning with distribution shift, even the choice of an appropriate error benchmark remains challenging. Testable learning with distribution shift (TDS) was introduced to develop computational foundations for this setting: using labeled training data and unlabeled test data, the learner must either produce a predictor with low test error or detect distribution shift. However, standard agnostic TDS guarantees allow a constant-factor amplification of the optimum error even when the training and test distributions coincide. They therefore do not directly recover the guarantee of classical agnostic learning. We introduce shift multiaccuracy, a stronger error benchmark inspired by multiaccuracy from algorithmic fairness. This yields a TDS formulation that becomes equivalent to multiaccuracy when there is no shift. Using a recent nontrivial reduction from agnostic learning to multiaccurate prediction from prior work, we obtain the first TDS framework that formally generalizes agnostic learning through an efficient reduction preserving the training marginal. The framework retains existing algorithmic upper bounds and inherits computational lower bounds from agnostic learning. Moreover, we establish a direct connection between our model and identity testing, yielding the first statistical characterization of agnostic TDS learning, in terms of Rademacher complexity. More broadly, we argue that shift multiaccuracy might be of independent interest in unsupervised domain adaptation, as it leads to sharper out-of-distribution generalization guarantees, and can be minimized directly, since its definition does not involve the unobserved test labels.
本条情报归属 AITOP「论文」核心栏目,重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」「中美大模型技术代差与商业化对照」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:multiaccuracy、分布偏移、agnostic learning、域适应,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.LG」(@Konstantinos Stavropoulos),经 AITOP 多维评分管线结构化提炼。