Exploring Hyperparameter Evolution in Machine Learning

Ken Stanley & Jim have a wide-ranging talk about one of his co-authored papers, "Designing neural networks through neuroevolution"... Ken Stanley & Jim have a wide-ranging talk about one of his co-authored papers, "Designing neural networks through neuroevolution". They cover neuroevolution dynamics, resistance to evolutionary thinking in AI, the evolutionary timescale, understanding genetic algorithms, neural networks & their role in neuroevolution, Ken's unifying NeuroEvolution of Augmenting Topologies (NEAT) method, scaling NEAT, the importance of diversity in innovation, novelty search, indirect encoding, Hybercube-based NeuroEvolution of Augmenting Topologies (HyperNEAT), co-evolutionary dynamics, the evolution of learning, and much more. Episode Transcript Ken's Website Designing neural networks through neuroevolution JRS: EP130 Ken Stanley on Why Greatness Cannot Be Planned Ken Stanley leads a research team at OpenAI on the challenge of open-endedness. He was previously Charles Millican Professor of Computer Science at the University of Central Florida and was also a co-founder of Geometric Intelligence Inc., which was acquired by Uber to create Uber AI Labs, where he was head of Core AI research.  He received a B.S.E. from the University of Pennsylvania in 1997 and received a Ph.D. in 2004 from the University of Texas at Austin. He is an inventor of the Neuroevolution of Augmenting Topologies (NEAT), HyperNEAT, novelty search, and POET algorithms, as well as the CPPN representation, among many others.