Serial vs. Parallel Turbo-Autoencoders and Accelerated Training for Learned Channel Codes.
Jannis ClausiusSebastian DörnerSebastian CammererStephan ten BrinkPublished in: ISTC (2021)
Keyphrases
- feedforward neural networks
- training set
- denoising
- training process
- parallel processing
- training examples
- real time
- restricted boltzmann machine
- online learning
- test set
- massively parallel
- parallel algorithm
- unsupervised learning
- supervised learning
- computer software
- training stage
- unsupervised manner
- learning stage
- deep belief networks
- training phase
- parallel implementation
- training algorithm
- training samples
- active learning
- evolutionary algorithm
- training data
- genetic algorithm