Using domain knowledge in coevolution and reinforcement learning to simulate a logistics enterprise.
Ying ZhaoErik HembergNate DerbinskyGabino MataUna-May O'ReillyPublished in: GECCO Companion (2022)
Keyphrases
- domain knowledge
- reinforcement learning
- temporal difference learning
- function approximation
- domain experts
- state space
- reinforcement learning algorithms
- supply chain management
- optimal policy
- information management
- domain ontology
- knowledge management
- supply chain
- dynamic programming
- prior knowledge
- learning process
- markov decision processes
- model free
- background knowledge
- optimal control
- learning algorithm
- machine learning
- temporal difference
- enterprise application
- prior domain knowledge
- logistics industry
- rfid technology
- data sets
- markov decision process
- game playing
- domain theory
- multi agent
- training data
- data mining