Autoregressive Policies for Continuous Control Deep Reinforcement Learning.
Dmytro KorenkevychA. Rupam MahmoodGautham VasanJames BergstraPublished in: IJCAI (2019)
Keyphrases
- autoregressive
- reinforcement learning
- control policies
- control policy
- optimal policy
- gaussian markov random field
- moving average
- non stationary
- action space
- random fields
- policy search
- autoregressive model
- optimal control
- fitted q iteration
- markov decision processes
- markov decision process
- information retrieval
- reward function
- random field models
- spectrum analysis
- autoregressive moving average
- sar images
- state space
- multiresolution
- frequency domain
- higher order
- semi supervised
- dynamic programming