RLBS: An Adaptive Backtracking Strategy Based on Reinforcement Learning for Combinatorial Optimization.
Ilyess BachiriJonathan GaudreaultClaude-Guy QuimperBrahim Chaib-draaPublished in: ICTAI (2015)
Keyphrases
- combinatorial optimization
- reinforcement learning
- combinatorial optimization problems
- traveling salesman problem
- branch and bound
- simulated annealing
- metaheuristic
- mathematical programming
- optimization problems
- branch and bound algorithm
- constraint satisfaction
- combinatorial problems
- memetic algorithm
- function approximation
- graph theory
- single machine scheduling problem
- quadratic assignment problem
- model free
- hard combinatorial optimization problems
- vehicle routing problem
- search strategy
- markov decision processes
- max flow min cut
- exact algorithms
- optimal policy
- search algorithm
- multi agent