RAPID-RL: A Reconfigurable Architecture with Preemptive-Exits for Efficient Deep-Reinforcement Learning.
Adarsh Kumar KostaMalik Aqeel AnwarPriyadarshini PandaArijit RaychowdhuryKaushik RoyPublished in: ICRA (2022)
Keyphrases
- reinforcement learning
- reconfigurable architecture
- function approximation
- markov decision processes
- temporal difference
- optimal policy
- state space
- reinforcement learning algorithms
- scheduling algorithm
- model free
- action space
- multi agent
- direct policy search
- learning process
- pattern recognition
- machine learning
- partially observable
- policy iteration
- rl algorithms
- continuous state
- autonomous learning
- reinforcement learning problems