富者愈富：半监督学习的差异性影响
The Rich Get Richer: Disparate Impact of Semi-Supervised Learning
AI总结：
本文揭示半监督学习对不同子群体产生差异性影响：基线准确率高的“富者”获益更多，而基线低的“贫者”可能性能下降，并提出收益比指标以促进公平性评估。
摘要：
半监督学习（SSL）在高质量监督数据严重受限时，已展现出提升多种学习任务模型准确率的潜力。尽管通常已证实整个数据群体的平均准确率得到提升，但尚不清楚SSL对不同子群体的表现如何。当不同子群体由我们旨在公平对待的人口统计群体定义时，理解上述问题具有重大的公平性影响。在本文中，我们揭示了部署SSL的差异性影响：未使用SSL时基线准确率较高的子群体（“富者”）往往从SSL中获益更多；而基线准确率较低的子群体（“贫者”）在添加SSL模块后甚至可能观察到性能下降。我们针对广泛类别的SSL算法，从理论和实证上确立了上述观察，这些算法要么显式要么隐式地使用辅助“伪标签”。在一系列图像和文本分类任务上的实验证实了我们的主张。我们引入了一个新指标，即收益比（Benefit Ratio），并促进对SSL公平性的评估（均衡收益比）。我们进一步讨论了如何缓解这种差异性影响。我们希望我们的论文能警示使用SSL的潜在陷阱，并鼓励对未来SSL算法进行多方面的评估。
英文摘要：
Semi-supervised learning (SSL) has demonstrated its potential to improve the model accuracy for a variety of learning tasks when the high-quality supervised data is severely limited. Although it is often established that the average accuracy for the entire population of data is improved, it is unclear how SSL fares with different sub-populations. Understanding the above question has substantial fairness implications when different sub-populations are defined by the demographic groups that we aim to treat fairly. In this paper, we reveal the disparate impacts of deploying SSL: the sub-population who has a higher baseline accuracy without using SSL (the "rich" one) tends to benefit more from SSL; while the sub-population who suffers from a low baseline accuracy (the "poor" one) might even observe a performance drop after adding the SSL module. We theoretically and empirically establish the above observation for a broad family of SSL algorithms, which either explicitly or implicitly use an auxiliary "pseudo-label". Experiments on a set of image and text classification tasks confirm our claims. We introduce a new metric, Benefit Ratio, and promote the evaluation of the fairness of SSL (Equalized Benefit Ratio). We further discuss how the disparate impact can be mitigated. We hope our paper will alarm the potential pitfall of using SSL and encourage a multifaceted evaluation of future SSL algorithms.
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。
