Abstract:MOTIVATION:While generative models have shown great success in sampling high-dimensional samples conditional on low-dimensional descriptors (stroke thickness in MNIST, hair color in CelebA, speaker identity in WaveNet), their generation out-of-distribution poses fundamental problems due to the difficulty of learning compact joint distribution across conditions. The canonical example of the conditional variational autoencoder (CVAE), for instance, does not explicitly relate conditions during training and, hence, has no explicit incentive of learning such a compact representation.RESULTS:We overcome the limitation of the CVAE by matching distributions across conditions using maximum mean discrepancy in the decoder layer that follows the bottleneck. This introduces a strong regularization both for reconstructing samples within the same condition and for transforming samples across conditions, resulting in much improved generalization. As this amount to solving a style-transfer problem, we refer to the model as transfer VAE (trVAE). Benchmarking trVAE on high-dimensional image and single-cell RNA-seq, we demonstrate higher robustness and higher accuracy than existing approaches. We also show qualitatively improved predictions by tackling previously problematic minority classes and multiple conditions in the context of cellular perturbation response to treatment and disease based on high-dimensional single-cell gene expression data. For generic tasks, we improve Pearson correlations of high-dimensional estimated means and variances with their ground truths from 0.89 to 0.97 and 0.75 to 0.87, respectively. We further demonstrate that trVAE learns cell-type-specific responses after perturbation and improves the prediction of most cell-type-specific genes by 65%.AVAILABILITY AND IMPLEMENTATION:The trVAE implementation is available via github.com/theislab/trvae. The results of this article can be reproduced via github.com/theislab/trvae_reproducibility.

Leveraging VQ-VAE tokenization for autoregressive modeling of medical time series

Generating Synthetic Mixed-Type Longitudinal Electronic Health Records for Artificial Intelligent Applications

Causal Recurrent Variational Autoencoder for Medical Time Series Generation

Synthetic electronic health records generated with variational graph autoencoders

Vector Quantized Time Series Generation with a Bidirectional Prior Model

Explainable Time Series Anomaly Detection using Masked Latent Generative Modeling

TimeVAE: A Variational Auto-Encoder for Multivariate Time Series Generation

Medical data wrangling with sequential variational autoencoders

Conditional Out-of-distribution Generation for Unpaired Data Using Transfer VAE.

Autoregressive Image Generation using Residual Quantization

Conditional out-of-sample generation for unpaired data using trVAE

Bayesian hierarchical vector autoregressive models for patient-level predictive modeling

Time-Causal VAE: Robust Financial Time Series Generator

Blending Low and High-Level Semantics of Time Series for Better Masked Time Series Generation

Deep Generative Modeling for Financial Time Series with Application in VaR: A Comparative Review

A Vector Quantized Variational Autoencoder (VQ-VAE) Autoregressive Neural $F_0$ Model for Statistical Parametric Speech Synthesis

Privacy-preserving datasets by capturing feature distributions with Conditional VAEs

RegaVAE: A Retrieval-Augmented Gaussian Mixture Variational Auto-Encoder for Language Modeling

Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs

Towards Accurate Image Coding: Improved Autoregressive Image Generation with Dynamic Vector Quantization