Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation
Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation
复制标题
DOI：
10.1109/cvpr42600.2020.00894
复制
发表时间：
2019-11
期刊：
影响因子：
--
通讯作者：
Zeyu Wang;Klint Qinami;Yannis Karakozis;Kyle Genova;P. Nair;K. Hata;Olga Russakovsky
中科院分区：
文献类型：
--
作者：
Zeyu Wang;Klint Qinami;Yannis Karakozis;Kyle Genova;P. Nair;K. Hata;Olga Russakovsky
计算机视觉模型通过从训练数据中获取相关统计数据来学习执行任务。研究表明，模特在接受活动识别或图像字幕等看似无关的任务时，会学习虚假的年龄、性别和种族关联。已经提出了各种缓解技术来防止模型利用或学习这种偏差。然而，这些技术之间几乎没有系统的比较。我们设计了一个简单但令人惊讶的有效视觉识别基准来研究偏差缓解。使用这个基准，我们提供了一系列技术的透彻分析。我们强调了目前流行的对抗训练方法的不足，提出了一种简单但同样有效的方法来替代赵等人的推理时间减少偏差放大方法，并设计了一种独立于领域的训练方法，其性能优于所有其他方法。最后，我们在CelebA数据集中验证了我们在属性分类任务上的发现，在CelebA数据集中，属性存在与图像中的人的性别相关，并证明了所提出的技术在缓解现实世界的性别偏见方面是有效的。
Computer vision models learn to perform a task by capturing relevant statistics from training data. It has been shown that models learn spurious age, gender, and race correlations when trained for seemingly unrelated tasks like activity recognition or image captioning. Various mitigation techniques have been presented to prevent models from utilizing or learning such biases. However, there has been little systematic comparison between these techniques. We design a simple but surprisingly effective visual recognition benchmark for studying bias mitigation. Using this benchmark, we provide a thorough analysis of a wide range of techniques. We highlight the shortcomings of popular adversarial training approaches for bias mitigation, propose a simple but similarly effective alternative to the inference-time Reducing Bias Amplification method of Zhao et al., and design a domain-independent training technique that outperforms all other methods. Finally, we validate our findings on the attribute classification task in the CelebA dataset, where attribute presence is known to be correlated with the gender of people in the image, and demonstrate that the proposed technique is effective at mitigating real-world gender bias.