论文标题
通过共同传播学习公正的表示
Learning Unbiased Representations via Mutual Information Backpropagation
论文作者
论文摘要
我们有兴趣学习数据驱动的表示，即使对固有的有偏见的数据进行了培训，这些表示驱动的表示。特别是，我们面临的情况是，如果模型学到的数据的某些属性（偏见）可以严重损害其泛化属性。我们通过信息理论的角度来解决这个问题，利用最新发现来对共同信息进行可区分的估计。我们提出了一种新颖的端到端优化策略，该策略同时估算并最大程度地减少了学习表示和数据属性之间的相互信息。当应用标准基准测试时，我们的模型在最先进的方法方面显示出可比或出色的分类性能。此外，我们的方法足够一般，可以适用于``算法公平性''的问题，并具有竞争性的结果。
We are interested in learning data-driven representations that can generalize well, even when trained on inherently biased data. In particular, we face the case where some attributes (bias) of the data, if learned by the model, can severely compromise its generalization properties. We tackle this problem through the lens of information theory, leveraging recent findings for a differentiable estimation of mutual information. We propose a novel end-to-end optimization strategy, which simultaneously estimates and minimizes the mutual information between the learned representation and the data attributes. When applied on standard benchmarks, our model shows comparable or superior classification performance with respect to state-of-the-art approaches. Moreover, our method is general enough to be applicable to the problem of ``algorithmic fairness'', with competitive results.
