Discrete Event Modeling and Simulation Aspects to Improve Machine Learning Systems

Abstract

Discrete Event Modeling and Simulation (M&S) and Machine Learning (ML) are two frameworks suited for system modeling which when combined can give powerful tools for system optimization for example. This paper details how discrete event M&S could be integrated into ML concepts and tools in order to improve the design and use of ML frameworks. An overview of different improvements are given and three concerning Reinforcement Learning (RL) are implemented in the framework of the DEVS formalism.