DCAST：多样化类别感知自训练缓解选择偏差以实现更公平的学习
DCAST: Diverse Class-Aware Self-Training Mitigates Selection Bias for Fairer Learning
AI总结：
针对机器学习中未归因于敏感特征的未识别偏差问题，提出感知类别特定偏差的DCAST自训练方法和层级偏差诱导方式，在11个数据集上验证了其提升模型公平性与鲁棒性的效果，多分类任务优势显著。
摘要：
机器学习中的公平性旨在缓解模型因性别或年龄等敏感特征对个体产生的偏见，这类偏见通常由选择偏差导致训练数据中人群分布不均所引发。值得注意的是，不归因于敏感特征的偏差难以识别，且通常未被诊断，尽管它在计算机视觉和分子生物医学等领域的复杂高维数据中十分突出。缓解未识别偏差以及评估缓解策略的方法至关重要，但目前仍未得到充分探索。我们提出：（i）多样化类别感知自训练（DCAST），这是一种感知类别特定偏差的模型无关缓解方法，它在利用无标签样本改善底层人群表征的同时，提升样本多样性以对抗传统自训练的确认偏差；（ii）层级偏差，一种无需先验知识的多变量类别感知偏差诱导方法。在11个数据集上，与传统自训练及6种主流领域自适应技术相比，使用DCAST训练的模型对层级偏差和其他偏差的鲁棒性均有所提升。该方法在多类别分类任务上优势最为显著，表明DCAST是一种适用于不同场景下更公平学习的有前景的策略。
英文摘要：
Fairness in machine learning seeks to mitigate model bias against individuals based on sensitive features such as sex or age, often caused by an uneven representation of the population in the training data due to selection bias. Notably, bias unascribed to sensitive features is challenging to identify and typically goes undiagnosed, despite its prominence in complex high-dimensional data from fields like computer vision and molecular biomedicine. Strategies to mitigate unidentified bias and evaluate mitigation methods are crucially needed, yet remain underexplored. We introduce: (i) Diverse Class-Aware Self-Training (DCAST), model-agnostic mitigation aware of class-specific bias, which promotes sample diversity to counter confirmation bias of conventional self-training while leveraging unlabeled samples for an improved representation of the underlying population; (ii) hierarchy bias, multivariate and class-aware bias induction without prior knowledge. Models learned with DCAST showed improved robustness to hierarchy and other biases across eleven datasets, against conventional self-training and six prominent domain adaptation techniques. Advantage was largest on multi-class classification, emphasizing DCAST as a promising strategy for fairer learning in different contexts.
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。
