Taming Normalizing Flows
This addresses privacy and bias issues in image generation, particularly for human faces, though it appears incremental as it applies existing fine-tuning techniques to Normalizing Flows.
The paper tackles the problem of controlling the output probability distribution of Normalizing Flow models for specific images or categories, achieving this through a fast fine-tuning process that takes minutes while maintaining generation quality.
We propose an algorithm for taming Normalizing Flow models - changing the probability that the model will produce a specific image or image category. We focus on Normalizing Flows because they can calculate the exact generation probability likelihood for a given image. We demonstrate taming using models that generate human faces, a subdomain with many interesting privacy and bias considerations. Our method can be used in the context of privacy, e.g., removing a specific person from the output of a model, and also in the context of debiasing by forcing a model to output specific image categories according to a given target distribution. Taming is achieved with a fast fine-tuning process without retraining the model from scratch, achieving the goal in a matter of minutes. We evaluate our method qualitatively and quantitatively, showing that the generation quality remains intact, while the desired changes are applied.