期刊
WIRELESS NETWORKS
卷 28, 期 3, 页码 1263-1267出版社
SPRINGER
DOI: 10.1007/s11276-018-01903-6
关键词
Deep learning; Data mining; Energy-based learning; Restricted Boltzmann machines
资金
- National Institutes of Health [GM062920]
Restricted Boltzmann machines are generative neural networks that can reconstruct missing data or classify new data. Continuous Boltzmann machines represent data with higher fidelity compared to discrete ones. The efficiency of the training algorithm is the primary limitation when using Boltzmann machines for big-data problems.
Restricted Boltzmann machines are a generative neural network. They summarize their input data to build a probabilistic model that can then be used to reconstruct missing data or to classify new data. Unlike discrete Boltzmann machines, where the data are mapped to the space of integers or bitstrings, continuous Boltzmann machines directly use floating point numbers and therefore represent the data with higher fidelity. The primary limitation in using Boltzmann machines for big-data problems is the efficiency of the training algorithm. This paper describes an efficient deterministic algorithm for training continuous machines.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据