Hybrid Learning Using Profit Sharing and Genetic Algorithm for Partially Observable Markov Decision Processes.
Kohei SuzukiShohei KatoPublished in: NBiS (2017)
Keyphrases
- hybrid learning
- partially observable markov decision processes
- profit sharing
- reinforcement learning
- genetic algorithm
- neuro fuzzy
- optimal policy
- finite state
- state space
- online learning
- dynamical systems
- markov decision processes
- belief state
- partially observable
- higher education
- supply chain
- joint replenishment
- dynamic programming
- learning algorithm
- neural network
- game theory
- decision problems
- blended learning
- markov decision process
- machine learning
- uncertain environments
- e learning
- multi agent
- learning process
- data mining
- optimal control
- infinite horizon
- long run
- case study
- belief revision
- objective function
- markov chain
- fuzzy logic