Martin Zaefferer
Papers
1
Total Citations
7
H-Index
1
About
Martin Zaefferer is a leading researcher in the field of surrogate-assisted optimization, with a particular focus on reducing the computational burden of expensive objective functions. His work centers on the development of phenotypic modeling techniques, which map genomes directly to performance outcomes, enabling efficient prediction of neural network behavior without costly evaluations. Zaefferer’s most cited paper, "Prediction of neural network performance by phenotypic modeling" (2019, 7 citations), introduces innovative surrogate models that replace traditional objective functions, significantly accelerating evolutionary optimization processes. This contribution is pivotal for applications where each evaluation is resource-intensive, such as in complex engineering design or hyperparameter tuning. Beyond this, Zaefferer has advanced the understanding of model-based optimization, bridging the gap between machine learning and evolutionary computation. His research is highly regarded for its practical impact, offering scalable solutions that reduce time and computational costs. For students and researchers, Zaefferer’s work provides essential tools for tackling real-world optimization challenges, making him a key figure in the evolution of efficient, data-driven problem-solving.
Research Focus
Key Achievements
Top Papers
- 1Prediction of neural network performance by phenotypic modeling7 citations · 2019