Overestimation, Overfitting, and Plasticity in Actor-Critic: the Bitter Lesson of Reinforcement Learning.
Michal NaumanMichal BortkiewiczMateusz OstaszewskiPiotr MilosTomasz TrzcinskiMarek CyganPublished in: CoRR (2024)
Keyphrases
- actor critic
- reinforcement learning
- temporal difference
- reinforcement learning algorithms
- optimal control
- policy gradient
- approximate dynamic programming
- gradient method
- neuro fuzzy
- policy iteration
- function approximation
- state space
- multi agent
- optimal policy
- model free
- feed forward
- average reward
- decision trees
- dynamic programming
- control problems
- temporal difference learning
- learning algorithm
- rl algorithms
- natural actor critic
- learning problems
- markov decision processes
- function approximators
- action selection
- planning problems
- fixed point
- training set
- neural network
- policy gradient methods