Alexander Hagg

Papers

Total Citations

H-Index

About

Dr. Alexander Hagg is a leading researcher in evolutionary computation and quality diversity optimization, with a focus on bridging the gap between algorithmic theory and real-world robotic design. His work centers on improving the efficiency and effectiveness of evolutionary algorithms through advanced surrogate modeling and representation analysis. In his highly cited 2019 paper, "Prediction of neural network performance by phenotypic modeling," Hagg introduced novel surrogate models that map genomes directly to objective values, significantly reducing the computational burden of evaluating expensive fitness functions in optimization tasks. This contribution has been pivotal for accelerating neural network design. More recently, his 2023 study, "On the Suitability of Representations for Quality Diversity Optimization of Shapes," critically examines how different encoding strategies impact the performance of quality diversity (QD) algorithms in robotic domains, revealing nuanced insights that challenge previous assumptions. With over 9 citations across his key works, Hagg’s research provides essential guidance for practitioners seeking to deploy QD methods in shape optimization and robotics, establishing him as a thoughtful voice in the evolution of representation learning and surrogate-assisted optimization.