Tackling the Credit Assignment Problem in Reinforcement Learning-Induced Pedagogical Policies with Neural Networks.
Markel Sanz AusinMehak ManiktalaTiffany BarnesMin ChiPublished in: AIED (1) (2021)
Keyphrases
- reinforcement learning
- neural network
- optimal policy
- policy search
- learning process
- markov decision process
- function approximators
- learning environment
- control policies
- state space
- pattern recognition
- markov decision processes
- function approximation
- e learning
- partially observable markov decision processes
- fitted q iteration
- back propagation
- credit scoring
- reinforcement learning agents
- reinforcement learning algorithms
- control policy
- learning objects
- fuzzy logic
- hierarchical reinforcement learning
- reward function
- credit risk
- markov decision problems
- learning theory
- dynamic programming
- recurrent neural networks
- learning activities
- machine learning
- artificial neural networks
- collaborative learning
- temporal difference
- multilayer perceptron
- multi agent
- multi agent reinforcement learning
- decision problems
- learning algorithm
- risk analysis
- learning capabilities
- genetic algorithm
- learning classifier systems
- policy gradient methods
- decentralized control
- transfer learning
- np hard
- neural nets
- feed forward
- action selection
- fuzzy systems
- learning design