HDPG: hyperdimensional policy-based reinforcement learning for continuous control.
Yang NiMariam IssaDanny AbrahamMahdi ImaniXunzhao YinMohsen ImaniPublished in: DAC (2022)
Keyphrases
- reinforcement learning
- control policy
- optimal policy
- control problems
- action space
- control policies
- action selection
- continuous state spaces
- control system
- optimal control
- markov decision process
- partially observable
- state space
- policy search
- markov decision problems
- robot control
- control strategies
- markov decision processes
- partially observable environments
- policy gradient
- continuous state
- policy evaluation
- long run
- function approximation
- decision problems
- reinforcement learning problems
- robotic control
- fitted q iteration
- reward function
- function approximators
- neural network
- adaptive control
- infinite horizon
- control method
- mobile robot
- multi agent
- machine learning