Gaussian Approximation for Bias Reduction in Q-Learning.
Carlo D'EramoAndrea CiniAlessandro NuaraMatteo PirottaCesare AlippiJan PetersMarcello RestelliPublished in: J. Mach. Learn. Res. (2021)
Keyphrases
- gaussian convolution
- linear computational complexity
- cooperative
- reinforcement learning
- state space
- series expansion
- multi agent
- polynomial approximation
- function approximation
- learning algorithm
- approximation error
- error bounds
- efficient computation
- action selection
- approximation algorithms
- gaussian noise
- multi agent reinforcement learning
- evaluation function
- model free
- queueing networks
- maximum likelihood
- special case
- multi agent systems
- multiscale
- machine learning
- taylor series expansion