RL-Sizer: VLSI Gate Sizing for Timing Optimization using Deep Reinforcement Learning.
Yi-Chen LuSiddhartha NathVishal KhandelwalSung Kyu LimPublished in: DAC (2021)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- optimal policy
- model free
- markov decision processes
- state space
- optimization algorithm
- action space
- signal processing
- temporal difference learning
- rl algorithms
- learning process
- multi agent
- autonomous learning
- power losses
- temporal difference
- control problems
- optimization problems
- supervised learning
- learning algorithm
- action selection
- direct policy search
- approximate dynamic programming
- partially observable domains
- dynamic programming
- markov decision problems
- continuous state
- machine learning
- reinforcement learning problems
- actor critic
- multi agent reinforcement learning
- control policy
- function approximators
- high speed
- partially observable
- learning capabilities
- learning classifier systems