Fast Population-Based Reinforcement Learning on a Single Machine.
Arthur FlajoletClaire Bizon MonrocKarim BeguirThomas PierrotPublished in: CoRR (2022)
Keyphrases
- single machine
- reinforcement learning
- scheduling problem
- dynamic programming
- maximum lateness
- earliness tardiness
- processing times
- scheduling jobs
- total weighted tardiness
- release dates
- total tardiness
- minimize total
- release times
- weighted number of tardy jobs
- optimal policy
- rolling horizon
- weighted tardiness
- sequence dependent setup times
- learning algorithm
- learning effect
- particle swarm optimization
- single machine scheduling problem
- state space
- optimal control
- simulated annealing
- competitive ratio
- machine learning
- production scheduling
- deteriorating jobs
- number of late jobs
- setup times
- differential evolution
- markov decision processes
- tabu search
- identical machines
- steady state