EC is a family of search methods that have been in widespread use since the 1960s. I will survey this history and main forms of evolutionary computation still in use. I will also introduce CPPN-NEAT, arguably the first generative AI method, which is also still in use due to its ease of implementation and robustness to hyper parameter settings. I will then describe a few use cases from my own group: symbolic regression, evolutionary robotics, and computationally designed organisms.
Back to Mathematics of Intelligences Tutorials