AlphaDDA: strategies for adjusting the playing strength of a fully trained AlphaZero system to a suitable human training partner.
Kazuhisa FujitaPublished in: PeerJ Comput. Sci. (2022)
Keyphrases
- training process
- training set
- human players
- supervised training
- test set
- human interaction
- human users
- multilayer perceptron
- feedforward neural networks
- multi layer perceptron
- discriminative training
- training algorithm
- motor skills
- contrastive divergence
- trained classifiers
- multilayer neural network
- data sets
- training stage
- serious games
- computer games
- semi supervised
- artificial neural networks
- training data
- neural network