MapZero: Mapping for Coarse-grained Reconfigurable Architectures with Reinforcement Learning and Monte-Carlo Tree Search.
Xiangyu KongYi HuangJianfeng ZhuXingchen ManYang LiuChunyang FengPengfei GouMinggui TangShaojun WeiLeibo LiuPublished in: ISCA (2023)
Keyphrases
- coarse grained
- monte carlo tree search
- fine grained
- reinforcement learning
- temporal difference
- bayesian reinforcement learning
- monte carlo
- reinforcement learning methods
- temporal difference learning
- evaluation function
- function approximation
- high level
- reinforcement learning algorithms
- state space
- optimal policy
- game tree
- protein sequences
- dynamic programming
- model free
- markov decision processes
- supervised learning
- action selection
- control problems
- transfer learning
- monte carlo search
- game playing
- optimal control
- search space
- policy iteration
- access control
- multi agent
- markov chain