FONDUE: an algorithm to find the optimal dimensionality of the latent representations of variational autoencoders
When training a variational autoencoder (VAE) on a given dataset, determining the optimal number of latent variables is mostly done by grid search: a costly process in terms of computational time and carbon footprint. In this paper, we explore the intrinsic dimension estimation (IDE) of the data and latent representations learned by VAEs. We show that the discrepancies between the IDE of the mean and sampled representations of a VAE after only a few steps of training reveal the presence of passive variables in the latent space, which, in well-behaved VAEs, indicates a superfluous number of dimensions. Using this property, we propose FONDUE: an algorithm which quickly finds the number of latent dimensions after which the mean and sampled representations start to diverge (i.e., when passive variables are introduced), providing a principled method for selecting the number of latent dimensions for VAEs and autoencoders.
Code (1)
Similar Papers 제목 키워드 기반
FONDUE: A Framework for Node Disambiguation Using Network Embeddings
Real-world data often presents itself in the form of a network. Examples include social networks, citation networks, biological networks, and knowledge graphs. In their simplest form, networks represent real-life entitie…
Knowledge GraphsNetwork EmbeddingQuestion AnsweringMaskAAE: Latent space optimization for Adversarial Auto-Encoders
The field of neural generative models is dominated by the highly successful Generative Adversarial Networks (GANs) despite their challenges, such as training instability and mode collapse. Auto-Encoders (AE) with regular…
Model Selection for Nonnegative Matrix Factorization by Support Union Recovery
Nonnegative matrix factorization (NMF) has been widely used in machine learning and signal processing because of its non-subtractive, part-based property which enhances interpretability. It is often assumed that the late…
Model SelectionSparsifying dimensionality reduction of PDE solution data with Bregman learning
Classical model reduction techniques project the governing equations onto a linear subspace of the original state space. More recent data-driven techniques use neural networks to enable nonlinear projections. Whilst thos…
compressed sensingDecoderDimensionality ReductionDiscLDA: Discriminative Learning for Dimensionality Reduction and Classification
Probabilistic topic models (and their extensions) have become popular as models of latent structures in collections of text documents or images. These models are usually treated as generative models and trained using max…
ClassificationDimensionality ReductionGeneral ClassificationSupervised dimensionality reduction+1