S-MFRL: Spiking Mean Field Reinforcement Learning for Dynamic Resource Allocation of D2D Networks.
Pei-Gen YeYuan-Gen WangWeixuan TangPublished in: IEEE Trans. Veh. Technol. (2023)
Keyphrases
- dynamic resource allocation
- reinforcement learning
- machine learning
- spiking neural networks
- resource allocation
- function approximation
- state space
- markov random field
- social networks
- spiking neurons
- bio inspired
- learning algorithm
- allocation policies
- bayesian inference
- optimal control
- closed form
- multi agent
- feed forward
- complex networks
- markov decision processes
- em algorithm
- learning rules
- reinforcement learning algorithms
- free energy
- spike trains
- graphical models
- image registration
- probabilistic model