Tractable loss function and color image generation of multinary restricted Boltzmann machine

2020/11/27 by Juno Hwang, Wonseok Hwang, Hwang, Juno +3
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2011.13509

NueRIPS 2020 DiffCVGP workshop paper

arxiv created 2020/11/27 · arxiv updated 2020/11/30

Abstract

The restricted Boltzmann machine (RBM) is a representative generative model based on the concept of statistical mechanics. In spite of the strong merit of interpretability, unavailability of backpropagation makes it less competitive than other generative models. Here we derive differentiable loss functions for both binary and multinary RBMs. Then we demonstrate their learnability and performance by generating colored face images.

Related

• Matrix Variate RBM and Its Applications
• Restricted Boltzmann Machines: Introduction and Review
• Deep Restricted Boltzmann Networks
• The Gaussian-Multinoulli Restricted Boltzmann Machine: A Potts Model Extension of the GRBM
• Matrix Variate RBM Model with Gaussian Distributions