Learning to Walk
Exploring the fascinating concept of AI learning locomotion, Mars shares how a bipedal form, when not constrained by energy conservation, might develop its own unique methods of movement, like a windmill-like run. The conversation delves into using Unity's ML agents framework to create simulations, emphasizing the potential for visualizing neural network learning through engaging and creative analogies, such as Rube Goldberg machines.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
SDS 591: Simulations and Synthetic Data for Machine Learning — with Mars Buttfield-Addison
Related Questions