An adaptive synchronization approach for weights of deep reinforcement learning.
S. Amirreza BadranMansoor RezghiPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- weighted sum
- reinforcement learning algorithms
- function approximation
- markov decision processes
- model free
- adaptive learning
- relative importance
- learning algorithm
- weighted average
- state space
- linear combination
- optimal policy
- robotic control
- data mining
- optimal weights
- learning problems
- evaluation function
- dynamic programming
- policy search
- case study
- decision trees