Joe Harrison

Volledige naam

J. Harrison

E-mail

Joe.Harrison@cwi.nl

Telefoon

+31 20 592 4014

Kamer

M278

Afdeling(en)

Evolutionary Intelligence

Functie(s)

PhD student

Biography

I am currently working on developing explainable machine learning methods for medicine. Specifically by using a combination of Genetic Programming and Deep Learning methods. Some of my interest are evolutionary deep learning, curriculum learning and life-long learning. In my spare time I make videos and posters using fractals, compression artefacts and various deep learning techniques.

Harrison, J., Virgolin, M., Alderliesten, T., & Bosman, P. (2023). Mini-Batching, Gradient-Clipping, first-versus second-order: What works in Gradient-Based coefficient optimisation for Symbolic Regression'. In Proceedings of the Genetic and Evolutionary Computation Conference Companion (pp. 1127–1136). doi:10.1145/3583131.3590368
Harrison, J., Alderliesten, T., & Bosman, P. (2022). Gene-pool Optimal Mixing in Cartesian Genetic Programming. In Proceedings of PPSN 2022 (pp. 19–32). doi:10.1007/978-3-031-14721-0_2

Evolutionary eXplainable Artificial Medical INtelligence Engine (EXAMINE)

Joe Harrison

Biography

Publications

Current projects with external funding

Joe Harrison

Biography

Publications

Current projects with external funding

Cookies