Transflow Learning: Repurposing Flow Models Without Retraining

Abstract

It is well known that deep generative models have a rich latent space, and that it is possible to smoothly manipulate their outputs by traversing this latent space. Recently, architectures have emerged that allow for more complex manipulations, such as making an image look as though it were from a different class, or painted in a certain style. These methods typically require large amounts of training in order to learn a single class of manipulations. We present Transflow Learning, a method for transforming a pre-trained generative model so that its outputs more closely resemble data that we provide afterwards. In contrast to previous methods, Transflow Learning does not require any training at all, and instead warps the probability distribution from which we sample latent vectors using Bayesian inference. Transflow Learning can be used to solve a wide variety of tasks, such as neural style transfer and few-shot classification.

References in corpus (5)

• Conditional Image Generation with PixelCNN Decoders
• Variational Inference with Normalizing Flows
• Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
• Few-Shot Unsupervised Image-to-Image Translation
• Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programs

Cited by in corpus (5)

• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis
• Few-Shot Adaptation of Generative Adversarial Networks
• RG-Flow: A hierarchical and explainable flow model based on renormalization group and sparse prior
• Kullback-Leibler Divergence-Based Out-of-Distribution Detection with Flow-Based Generative Models