Learning with Training Wheels: Speeding up Training with a Simple Controller for Deep Reinforcement Learning.
Linhai XieSen WangStefano RosaAndrew MarkhamNiki TrigoniPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- supervised learning
- deep architectures
- training process
- learning process
- learning speed
- online learning
- online training
- computer based training
- training examples
- feedforward neural networks
- training set
- training program
- learning machines
- learning tasks
- learning problems
- structured prediction
- recurrent networks
- learning algorithm
- training data
- computer software
- learning stage
- skill acquisition
- motor skills
- action selection
- control rules
- learning capabilities
- training phase
- machine learning
- serious games
- control scheme
- optimal control
- function approximation
- markov decision processes
- learning environment