Reinforcement Learning for Cross-Domain Hyper-Heuristics.
Florian MischekNysret MusliuPublished in: IJCAI (2022)
Keyphrases
- cross domain
- hyper heuristics
- reinforcement learning
- transfer learning
- evolutionary algorithm
- genetic programming
- examination timetabling
- timetabling problem
- knowledge transfer
- cutting stock problems
- domain adaptation
- heuristic search
- difficult problems
- metaheuristic
- constraint satisfaction problems
- state space
- condition action rules
- cross domain learning
- search procedure
- target domain
- multiple domains
- text categorization
- learning algorithm
- learning tasks
- machine learning
- heuristic methods
- labeled data
- multi objective
- information retrieval
- search heuristics
- neural network
- fitness function
- genetic algorithm
- search space
- unlabeled data
- evolutionary computation