PrefixRL: Optimization of Parallel Prefix Circuits using Deep Reinforcement Learning.
Rajarshi RoyJonathan RaimanNeel KantIlyas ElkinRobert KirbyMichael Y. SiuStuart F. ObermanSaad GodilBryan CatanzaroPublished in: DAC (2021)
Keyphrases
- reinforcement learning
- optimization problems
- optimization algorithm
- parallel implementation
- high speed
- state space
- data structure
- learning algorithm
- machine learning
- optimal policy
- reinforcement learning algorithms
- parallel processing
- action selection
- function approximation
- constrained optimization
- temporal difference
- robotic control
- optimization methods
- global optimization
- markov decision processes
- mobile robot