Revisiting Scalable Hessian Diagonal Approximations for Applications in Reinforcement Learning.
Mohamed ElsayedHomayoon FarrahiFelix DangelA. Rupam MahmoodPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- policy evaluation
- machine learning
- learning process
- optimal control
- state space
- markov decision processes
- neural network
- reinforcement learning algorithms
- step size
- multi agent
- learning algorithm
- model free
- action selection
- data sets
- reinforcement learning methods
- temporal difference learning
- hessian matrix
- computationally tractable
- action space
- policy iteration
- web scale
- partially observable
- efficient computation
- learning classifier systems
- dynamical systems
- lightweight