Nonlinear Concept Erasure: a Density Matching Approach

Ensuring that neural models used in real-world applications cannot infer sensitive information, such as demographic attributes like gender or race, from text representations is a critical challenge when fairness is a concern. We address this issue through concept erasure, a process that removes information related to a specific concept from distributed representations while preserving as much of the remaining semantic information as possible. Our approach involves learning an orthogonal projection in the embedding space, designed to make the class-conditional feature distributions of the discrete concept to erase indistinguishable after projection. By adjusting the rank of the projector, we control the extent of information removal, while its orthogonality ensures strict preservation of the local structure of the embeddings. Our method, termed overlineLEOPARD, achieves state-of-the-art performance in nonlinear erasure of a discrete attribute on classic natural language processing benchmarks. Furthermore, we demonstrate that overlineLEOPARD effectively mitigates bias in deep nonlinear classifiers, thereby promoting fairness.