Monte Carlo simulation based algorithm design for automatic learning machine performance analysis

Abstract

This paper proposes advanced key performance indicators dedicated to learning machines and auto-adaptive systems performance analysis. This work introduces an algorithm implementing designed key performance indicators for automatic learning capacity checking. The algorithm simulates the supervised environment to stimulate the tested auto-adaptive machine and then study its adaption capacity based on indicators statistically designed.