Dynamic Narrowing of VAE Bottlenecks Using GECO and L0 Regularization

Abstract

When designing variational autoencoders (VAEs) or other types of latent space models, the dimensionality of the latent space is typically defined upfront. In this process, it is possible that the number of dimensions is under- or overprovisioned for the application at hand. In case the dimensionality is not predefined, this parameter is usually determined using time- and resource-consuming cross-validation. For these reasons we have developed a technique to shrink the latent space dimensionality of VAEs automatically and on-the-fly during training using Generalized ELBO with Constrained Optimization (GECO) and the $L_0$-Augment-REINFORCE-Merge ($L_0$-ARM) gradient estimator. The GECO optimizer ensures that we are not violating a predefined upper bound on the reconstruction error. This paper presents the algorithmic details of our method along with experimental results on five different datasets. We find that our training procedure is stable and that the latent space can be pruned effectively without violating the GECO constraints.

arXiv comment8 pages, 3 figures, 1 table

References in corpus (11)

• Efficient Estimation of Word Representations in Vector Space
• Auto-Encoding Variational Bayes
• Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
• Rethinking the Value of Network Pruning
• Deep Variational Information Bottleneck
• Variational Dropout Sparsifies Deep Neural Networks
• Compressing Neural Networks using the Variational Information Bottleneck
• ARM: Augment-REINFORCE-Merge Gradient for Stochastic Binary Networks
• Taming VAEs
• $L_0$-ARM: Network Sparsification via Stochastic Binary Optimization
• Stochastic Variational Optimization