NeurIPS 2020

Learning Certified Individually Fair Representations

Meta Review

The reviewing committee agrees this is a strong proposal that borrows tools from recent advances in constraint satisfaction to address a critical challenge in ML fairness (certification for individually fair representation). The paper is clearly written and the experiments are well conducted. We recommend its acceptance for publication and suggest that authors incorporate in the final version the clarifications and results included in the rebuttal. The authors are also encourage to include additional baselines in the final version of the paper.