Quantitative Universal Approximation Bounds for Deep Belief Networks

2022/08/18 by Julian Sieber, Sieber, Julian, Johann Gehringer +1
Computer Science · #Adversarial Robustness in Machine Learning #Domain Adaptation and Few-Shot Learning #Machine Learning and Algorithms

paper · pdf · doi:10.48550/arxiv.2208.09033

Abstract

We show that deep belief networks with binary hidden units can approximate any multivariate probability density under very mild integrability requirements on the parental density of the visible nodes. The approximation is measured in the L^q-norm for q∈[1,∞] (q=∞ corresponding to the supremum norm) and in Kullback-Leibler divergence. Furthermore, we establish sharp quantitative bounds on the approximation error in terms of the number of hidden units.

Related

• Universal Approximation Depth and Errors of Narrow Belief Networks with Discrete Units
• Refinements of Universal Approximation Results for Deep Belief Networks and Restricted Boltzmann Machines
• Large Deviation Methods for Approximate Probabilistic Inference
• Kernels and Submodels of Deep Belief Networks
• A Universal Approximation Theorem of Deep Neural Networks for Expressing Probability Distributions