ContextFlow++: Generalist-Specialist Flow-based Generative Models with Mixed-Variable Context Encoding
This work addresses a practical limitation in flow-based models for applications requiring exact density estimation, offering an incremental improvement with specific gains in training stability and performance.
The paper tackles the problem of conditioning normalizing flow-based generative models on context, which is limited by conventional methods, and introduces ContextFlow++ with additive conditioning and mixed-variable context encoding to enable generalist-specialist setups, achieving faster stable training and higher performance metrics on benchmarks like rotated MNIST-R and corrupted CIFAR-10C.
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.