Uncertainty Autoencoders: Learning Compressed Representations via Variational Information Maximization

Abstract

Compressed sensing techniques enable efficient acquisition and recovery of sparse, high-dimensional data signals via low-dimensional projections. In this work, we propose Uncertainty Autoencoders, a learning framework for unsupervised representation learning inspired by compressed sensing. We treat the low-dimensional projections as noisy latent representations of an autoencoder and directly learn both the acquisition (i.e., encoding) and amortized recovery (i.e., decoding) procedures. Our learning objective optimizes for a tractable variational lower bound to the mutual information between the datapoints and the latent representations. We show how our framework provides a unified treatment to several lines of research in dimensionality reduction, compressed sensing, and generative modeling. Empirically, we demonstrate a 32% improvement on average over competing approaches for the task of statistical compressed sensing of high-dimensional datasets.

arXiv commentAISTATS 2019

Cited by in corpus (11)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Compressed Sensing with Deep Image Prior and Learned Regularization
• Deep-learning-based reduced-order modeling for subsurface flow simulation
• Inverting Deep Generative models, One layer at a time
• Deep Learning Techniques for Compressive Sensing-Based Reconstruction and Inference -- A Ubiquitous Systems Perspective
• Infomax Neural Joint Source-Channel Coding via Adversarial Bit Flip
• On the Generative Utility of Cyclic Conditionals
• Label-Based Diversity Measure Among Hidden Units of Deep Neural Networks: A Regularization Method
• Compressed Sensing via Measurement-Conditional Generative Models
• Representation Evaluation Block-based Teacher-Student Network for the Industrial Quality-relevant Performance Modeling and Monitoring
• Eccentric Regularization: Minimizing Hyperspherical Energy without explicit projection