Langevin-gradient parallel tempering for Bayesian neural learning.
Rohitash ChandraKonark JainRatneel Vikash DeoSally CrippsPublished in: CoRR (2018)
Keyphrases
- neural learning
- bayesian inference
- markov chain monte carlo
- multi layered
- probabilistic model
- prior information
- neural network
- restricted boltzmann machine
- hyperparameters
- incremental learning
- posterior probability
- posterior distribution
- particle filter
- markov chain
- parameter estimation
- generative model
- hidden variables
- reinforcement learning
- latent variables
- higher order
- gradient method
- machine learning