Debiasing Pretrained Generative Models by Uniformly Sampling Semantic Attributes

Part of Advances in Neural Information Processing Systems 36 (NeurIPS 2023) Main Conference Track

Bibtex Paper Supplemental


Walter Gerych, Kevin Hickey, Luke Buquicchio, Kavin Chandrasekaran, Abdulaziz Alajaji, Elke A. Rundensteiner, Emmanuel Agu


Generative models are being increasingly used in science and industry applications. Unfortunately, they often perpetuate the biases present in their training sets, such as societal biases causing certain groups to be underrepresented in the data. For instance, image generators may overwhelmingly produce images of white people due to few non-white samples in their training data. It is imperative to debias generative models so they synthesize an equal number of instances for each group, while not requiring retraining of the model to avoid prohibitive expense. We thus propose a distribution mapping module that produces samples from a fair noise distribution, such that the pretrained generative model produces semantically uniform outputs - an equal number of instances for each group - when conditioned on these samples. This does not involve retraining the generator, nor does it require any real training data. Experiments on debiasing generators trained on popular real-world datasets show that our method outperforms existing approaches.