Variational Autoencoder with Arbitrary Conditioning

Abstract

We propose a single neural probabilistic model based on variational autoencoder that can be conditioned on an arbitrary subset of observed features and then sample the remaining features in "one shot". The features may be both real-valued and categorical. Training of the model is performed by stochastic variational Bayes. The experimental evaluation on synthetic data, as well as feature imputation and image inpainting problems, shows the effectiveness of the proposed approach and diversity of the generated samples.

arXiv commentPublished as a conference paper at ICLR 2019

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• Conditional Generative Adversarial Nets
• MissForest - nonparametric missing value imputation for mixed-type data
• GAIN: Missing Data Imputation using Generative Adversarial Nets
• MADE: Masked Autoencoder for Distribution Estimation
• Data Generation as Sequential Decision Making
• A Universal Marginalizer for Amortized Inference in Generative Models
• Learning to Generate Samples from Noise through Infusion Training

Cited by in corpus (22)

• Learning Model-Agnostic Counterfactual Explanations for Tabular Data
• ControlVAE: Controllable Variational Autoencoder
• Learning Neural Causal Models from Unknown Interventions
• Networked Time Series Imputation via Position-aware Graph Enhanced Variational Autoencoders
• Variational Dynamic for Self-Supervised Exploration in Deep Reinforcement Learning
• not-MIWAE: Deep Generative Modelling with Missing not at Random Data
• Electra: Conditional Generative Model based Predicate-Aware Query Approximation
• InfoColorizer: Interactive Recommendation of Color Palettes for Infographics
• Active Feature Acquisition with Generative Surrogate Models
• Learning about an exponential amount of conditional distributions
• Classify and Generate: Using Classification Latent Space Representations for Image Generations
• Sage: Using Unsupervised Learning for Scalable Performance Debugging in Microservices
• Controllable and Diverse Text Generation in E-commerce
• Efficient Neural Causal Discovery without Acyclicity Constraints
• Variational Auto-Decoder: A Method for Neural Generative Modeling from Incomplete Data
• Monte Carlo guided Diffusion for Bayesian linear inverse problems
• δ-CLUE: Diverse Sets of Explanations for Uncertainty Estimates
• Arbitrary Conditional Distributions with Energy
• Joint data imputation and mechanistic modelling for simulating heart-brain interactions in incomplete datasets
• Projected Latent Markov Chain Monte Carlo: Conditional Sampling of Normalizing Flows
• Explainability as statistical inference
• VConstruct: Filling Gaps in Chl-a Data Using a Variational Autoencoder