Boosting Reinforcement Learning Algorithms in Continuous Robotic Reaching Tasks using Adaptive Potential Functions.
Yifei ChenLambert SchomakerFrancisco CruzPublished in: CoRR (2024)
Keyphrases
- reinforcement learning algorithms
- potential functions
- reinforcement learning
- learning algorithm
- state space
- weighted sums
- model free
- tabula rasa
- reinforcement learning problems
- markov decision processes
- markov random field
- reinforcement learning methods
- function approximation
- temporal difference
- maximum entropy
- reward function
- learning tasks
- evaluation function
- step size
- linear combination
- dynamic environments
- action space
- mobile robot
- feature selection
- machine learning