IEEE Members: Free
Non-members: FreeDuration: 01:04:04
Risto Miikkulainen (U. Texas at Austin, USA), Abstract:Neural network weights and topologies were originally evolved in order to solve tasks where gradients are not available. Recently, it has also become a useful technique for metalearning architectures of deep learning networks. However, neuroevolution is most powerful when it utilizes synergies of evolution and learning. In this talk I review four examples of such synergies: evolving loss functions, co-adapting learning and evolution, evolving activation functions, and evolving decision-making based on surrogate learning. I will demonstrate these synergies in image recognition, game playing, and pandemic policy optimization, and point out opportunities for future work.