A Formal Verification Model for Performance Analysis of Reinforcement Learning Algorithms Applied to Dynamic Networks
Journal Title: Journal of Applied Computer Science & Mathematics - Year 2017, Vol 11, Issue 23
Abstract
Routing data packets in a dynamic network is a difficult and important problem in computer networks. As the network is dynamic, it is subject to frequent topology changes and is subject to variable link costs due to congestion and bandwidth. Existing shortest path algorithms fail to converge to better solutions under dynamic network conditions. Reinforcement learning algorithms posses better adaptation techniques in dynamic environments. In this paper we apply model based Q-Routing technique for routing in dynamic network. To analyze the correctness of Q-Routing algorithms mathematically, we provide a proof and also implement a SPIN based verification model. We also perform simulation based analysis of Q-Routing for given metrics.
Authors and Affiliations
ALNASER As'ad Mahmoud As'ad
Keywords
Related Articles
- EP ID EP446671
- DOI 10.4316/JACSM.201701001
- Views 187
- Downloads 0
Complexities in a Plant-Herbivore Model
A simple host-parasite type model has been considered to study the interaction of certain plants and herbivores. The two dimensional discrete time model utilizes leaf and herbivore biomass as state variables. The paramet...
Real World Applications of MGR, Neeva and KN-Hash
Hash functions have prominent role in cryptography because of their ubiquitous applications in real world. Earlier, it was used for authentication only but with continuous research and development, it has been started us...
Approximate solution of high-order integro-differential equations using radial basis functions
In this paper, we present a numerical method to solve linear and nonlinear high-order Volterra integro-differential equations. This method is based on interpolating by radial basis functions, using Legendre-Gauss-Lobatto...
Application of Stand-PSO Technique for Optimization Cameras’ 2D Dispositions in a MoCap system
In this paper, a detailed study of the Particle Swarm Optimization (PSO) technique is given in its standard version to solve a network camera placement problem and to ensure the coverage of a reflector point by, at least...
Kalman Filters for Estimating the potential GDP
The estimation of the potential GDP has a twofold importance: on one hand its accurate estimation allows the correct dimensioning of the macroeconomic policies and on the other hand, the study of potential GDP is a resea...