AutoPhase: Compiler Phase-Ordering for HLS with Deep Reinforcement Learning.
Qijing HuangAmeer Haj-AliWilliam S. MosesJohn XiangIon StoicaKrste AsanovicJohn WawrzynekPublished in: FCCM (2019)
Keyphrases
- reinforcement learning
- multi agent
- programming language
- function approximation
- general purpose
- learning algorithm
- reinforcement learning algorithms
- markov decision processes
- software systems
- partial order
- learning process
- model free
- preprocessing phase
- multi agent reinforcement learning
- robotic control
- supervised learning
- monte carlo
- state space
- training phase
- action selection
- database
- temporal difference learning