Efficiently Evolving Algorithms for Real-World Applications
Date:
Abstract
What if evolutionary computation is not just a tool for optimization, but a design engine for algorithms themselves?
In this keynote, I argue that we are witnessing a shift from hand-crafted metaheuristics to automated algorithm construction, with evolutionary principles providing the backbone and Large Language Models and other generative AI techniques acting as high-level variation operators. Rather than replacing EC, LLMs amplify its expressive power, enabling search directly in program space.
Using the LLaMEA framework and its extensions (in-the-loop HPO, MAP-elites, Novelty search and others), I will show how evolutionary loops can generate competitive algorithms across continuous, combinatorial, and real-world engineering problems.
But how can these computationally expensive frameworks be employed for solving real-world problems efficiently? And how to perform rigorous benchmarking while using stochastic ever-changing LLMs?
The key question is no longer whether we can automate algorithm design, but how to better control, understand and utilize it efficiently to new domains.
Speaker Bio
Niki van Stein is Associate Professor at the Leiden Institute of Advanced Computer Science (LIACS), Leiden University, where she leads the XAI research group.
Her research sits at the intersection of explainable AI, automated algorithm design and metaheuristic optimization, with a particular focus on using large language models to automatically discover and improve optimization algorithms.
She is the lead developer of LLaMEA, an award-winning framework for LLM-driven algorithm evolution, and has broad experience applying these methods to real-world problems in engineering, predictive maintenance, and scientific design.
Alongside her research, she is active in the international evolutionary computation and AutoML communities as a conference organizer, program committee member and editorial board member. She holds a PhD from Leiden University and previously co-founded and led a software company as CTO, giving her a strong background in bridging academic research and industrial practice.