Great conversations when presenting my poster on the Weight Transformer MetaNCA at #ALife 2025! It's about learning local update rules for neural networks in order to be able to generate weights for arbitrary architectures ("architecture generalization").

Within the next 5 years there's going to be a better, more efficient way to train neural networks than gradient-based optimization. My bet is that it's going to take inspiration from living organisms, with properties like self-organization, replication, and evolution.