Creative Analogies For The Underpinnings of Deep Learning
no recording on file
summary
Explains why deep neural networks readily learn patterns like symmetry, locality, compositionality, and nonlinearity by using meta-mathematical analogies and concrete demonstrations. Uses practical GPU-backed code examples to show how those principles emerge in modern deep learning models.
AI-written summary · about
related talks
27mDemystifying Deep Neural Networks
34mThe Linear Algebra of Deep Learning
19mDeep Learning is Just LEGO: & Other Hands on Machine Learning Activities
32mHow Neuroscience Makes AI Intelligent
54mIntroduction to Understanding Deep Learning
24mSuper Accessible No Math Intro to Neural Networks For Beginners
51mDeep Learning - What is it and What It Can Do For You
44mLearning from Machines