M32.2 CONNECT THE MECHANISM
Evolve programs, weights, or architectures
Evolution can breed formulas, grow neural networks connection by connection, train robot walkers without backpropagation, and, paired with a language model, beat a matrix-multiplication record from 1969.
LESSON OVERVIEW14 min lesson
Lesson overview
Evolution can breed formulas, grow neural networks connection by connection, train robot walkers without backpropagation, and, paired with a language model, beat a matrix-multiplication record from 1969.
What you’ll explore
- Genetic programming searches executable structures, neuroevolution searches neural parameters or topology, and architecture search must separate candidate selection from independent evaluation.
GO TO THE SOURCE
Original explanations, connected to the research.
Genetic Programming: On the Programming of Computers by Means of Natural Selection (Koza, 1992)A Field Guide to Genetic Programming (Poli, Langdon & McPhee, 2008)Distilling Free-Form Natural Laws from Experimental Data (Schmidt & Lipson, 2009)Evolving Neural Networks through Augmenting Topologies (Stanley & Miikkulainen, 2002)Evolution Strategies as a Scalable Alternative to Reinforcement Learning (Salimans et al., 2017)Mathematical discoveries from program search with large language models (FunSearch; Romera-Paredes et al., 2023)AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms (Google DeepMind, 2025)Regularized Evolution for Image Classifier Architecture Search (Real et al., 2019)Suggest a correction
A precise note can make an explanation better.
Choose the scene and describe what needs attention. Download a feedback file to share through a channel you already use. This page does not send feedback or connect you with a reviewer.