Welcome to the Computational Intelligence Lab at the University of Oldenburg.

Evolution Strategies

Evolution Strategies (ES) are a central research topic of the Computational Intelligence Lab. The lab studies population-based optimization methods for complex, noisy, multimodal, and black-box problems. Research topics include self-adaptation, covariance matrix adaptation, CMA-ES, natural evolution strategies, multi-objective optimization, and constraint handling.

A current application area is prompt optimization for large language models. In this context, Evolution Strategies are used to automatically improve prompts, system contexts, configurations, and agentic workflows. The goal is to develop more powerful and robust AI systems, for example through optimized RAG-based extensions or ES-specific support agents for modeling, implementation, and analysis tasks.

Another application area is molecule optimization for drug design, where evolutionary search can support the development of candidate molecules with desirable properties such as binding affinity, stability, synthesizability, selectivity, and low toxicity.

The long-term goal is to develop efficient and reliable evolutionary optimization methods for intelligent systems, scientific applications, and AI-assisted algorithm design.

New Book "Artificial Intelligence Essentials", Springer, 2026

The book provides a structured introduction to core areas of artificial intelligence, including machine learning, deep learning, large language models, and agent-based systems. It combines theoretical foundations with practical implementation, using tools such as scikit-learn, Keras, and Ollama. Each chapter includes code examples, exercises, and links to interactive notebooks. The focus lies on conveying applicable knowledge for real-world use. The book is intended for students and practitioners in computer science or related fields and can be used for teaching, self-study, or reference in applied contexts.

Books

Blogposts on Towards Data Science

SPIEGEL Podcast - Moreno+1: »Eine starke KI? Halte ich für wahrscheinlich«

AI professor Oliver Kramer works on algorithms that exhibit “human-like cognitive performance.” He is convinced that machines will become more intelligent than humans. We should think about that.

Link to the podcast.