---
title: "On the Impact of Crossover in Many-Objective Optimization: A Runtime Analysis of NSGA-III"
authors:
  - "Andre Opris"
doi: "10.24963/ijcai.2026/705"
url: "https://ijcai.org/proceedings/2026/705"
pdf: "https://ijcai.org/proceedings/2026/0705.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "6334-6342"
track: "Main Track"
keywords:
  - "Search: Evolutionary computation"
  - "Search: Heuristic search"
---

# On the Impact of Crossover in Many-Objective Optimization: A Runtime Analysis of NSGA-III

Andre Opris

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 6334-6342. DOI: [10.24963/ijcai.2026/705](https://doi.org/10.24963/ijcai.2026/705)

## Abstract

In recent years, theoretical understanding has rapidly advanced regarding how popular multi-objective evolutionary algorithms (MOEAs) can optimize many-objective problems. However, the benefits of using crossover in many-objective optimization are theoretically not understood, except for specifically designed benchmark functions tuned to particular crossover operators, and still lag significantly behind its practical use. In this paper, we build upon this line of research and present a theoretical runtime analysis of the widely used NSGA-III algorithm on the classical m-objective m-OneJumpZeroJump function (m-OJZJ for short). Our results demonstrate that NSGA-III with crossover optimizes m-OJZJ asymptotically faster than NSGA-III without crossover for any number m of objectives for huge parameter regimes. We complement our analysis by providing a lower runtime bound on 4-OJZJ when crossover is turned off.