ContextFlow++: Generalist-Specialist Flow-based Generative Models with Mixed-Variable Context Encoding
Abstract
Normalizing flow-based generative models have been widely used in applications where the exact density estimation is of major importance. Recent research proposes numerous methods to improve their expressivity. However, conditioning on a context is largely overlooked area in the bijective flow research. Conventional conditioning with the vector concatenation is limited to only a few flow types. More importantly, this approach cannot support a practical setup where a set of context-conditioned (specialist) models are trained with the fixed pretrained general-knowledge (generalist) model. We propose ContextFlow++ approach to overcome these limitations using an additive conditioning with explicit generalist-specialist knowledge decoupling. Furthermore, we support discrete contexts by the proposed mixed-variable architecture with context encoders. Particularly, our context encoder for discrete variables is a surjective flow from which the context-conditioned continuous variables are sampled. Our experiments on rotated MNIST-R, corrupted CIFAR-10C, real-world ATM predictive maintenance and SMAP unsupervised anomaly detection benchmarks show that the proposed ContextFlow++ offers faster stable training and achieves higher performance metrics. Our code is publicly available at https://github.com/gudovskiy/contextflow.
Keywords
Cite
@article{arxiv.2406.00578,
title = {ContextFlow++: Generalist-Specialist Flow-based Generative Models with Mixed-Variable Context Encoding},
author = {Denis Gudovskiy and Tomoyuki Okuno and Yohei Nakata},
journal= {arXiv preprint arXiv:2406.00578},
year = {2024}
}
Comments
Accepted to UAI 2024. Preprint