Elizabeth Warner

Papers

Total Citations

H-Index

About

Elizabeth Warner is a foundational figure in the geometry of artificial intelligence benchmarking, best known for her meticulous work on the Tartarus fitness landscape. Her primary research areas include evolutionary computation, fitness case analysis, and the rigorous validation of AI task environments. Warner’s major contribution lies in her 2008 paper, "The geometry of Tartarus fitness cases," which systematically dissected the standard Tartarus grid-robot problem. She revealed that of the 320,320 possible fitness cases, only 297,040 are valid under the original problem statement—a critical correction that reshaped how researchers evaluate algorithmic performance. This work, cited 12 times, is a cornerstone for those studying deceptive fitness landscapes and benchmark integrity. Warner’s precision in mapping the geometry of these cases has provided a clearer, more reliable foundation for comparing evolutionary strategies. Her achievements underscore the importance of methodological rigor in AI, making her a respected authority for students and researchers navigating the complexities of fitness function design.