Niching Importance Sampling for Multi-modal Rare-event Simulation

TL;DR

This paper introduces niching importance sampling, a robust method combining reliability analysis and niching techniques to improve rare-event probability estimation in multi-modal problems.

Contribution

It presents a novel framework that enhances rare-event simulation by integrating niching with importance sampling and cross entropy minimization.

Findings

Consistently avoids degenerate behavior in multi-modal problems

Successfully applied to various numerical examples

Provides a robust estimator for failure probability

Abstract

This paper proposes niching importance sampling, a framework that combines concepts from reliability analysis, e.g. Markov chains, importance sampling, and relative cross entropy minimisation, with niching techniques from evolutionary multi-modal optimisation. The result is a highly robust estimator of the probability of failure, that can tackle sampling challenges posed by the underlying geometry of a reliability problem. Niching importance sampling is tested on a range of numerical examples and is shown to consistently avoid the degenerate behaviour observed for existing reliability methods on several multi-modal performance functions.