Entropy, Free Energy, and Work of Restricted Boltzmann Machines
A restricted Boltzmann machine is a generative probabilistic graphic network. A probability of finding the network in a certain configuration is given by the Boltzmann distribution. Given training data, its learning is done by optimizing the parameters of the energy function of the network. In this...
Main Authors: | Sangchul Oh, Abdelkader Baggag, Hyunchul Nha |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-05-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/22/5/538 |
Similar Items
-
Analysis on Noisy Boltzmann Machines and Noisy Restricted Boltzmann Machines
by: Wenhao Lu, et al.
Published: (2021-01-01) -
Dynamic Topology Reconfiguration of Boltzmann Machines on Quantum Annealers
by: Jeremy Liu, et al.
Published: (2020-10-01) -
Fine‐tuning restricted Boltzmann machines using quaternions and its application for spam detection
by: Luis A. daSilva, et al.
Published: (2019-05-01) -
belg: A Tool for Calculating Boltzmann Entropy of Landscape Gradients
by: Jakub Nowosad, et al.
Published: (2020-08-01) -
Boltzmann-Gibbs entropy: axiomatic characterization and application
by: C. G. Chakrabarti, et al.
Published: (2000-01-01)