Certified Defense to Image Transformations via Randomized Smoothing

Fair representation learning provides an effective way of enforcing fairness constraints without compromising utility for downstream users. A desirable family of such fairness constraints, each requiring similar treatment for similar individuals, is known as individual fairness. In this work, we introduce the first method thatenables data consumers to obtain certificates of individualfairness for existing andnew data points. The key idea is to map similar individuals toclose latent representations and leverage this latent proximity to certify individual fairness. That is, our method enables the data producer to learn and certify a representation where for a data point all similar individuals are at ℓ∞-distance at most epsilon, thus allowing data consumers to certify individual fairness by proving epsilon-robustness of their classifier. Our experimental evaluation on five real-world datasets and several fairnessconstraints demonstrates the expressivity and scalability of our approach