On the Connection between Neural Networks and Kernels: a Modern Perspective -Simon Du Published 2019-10-16 Download video MP4 360p Recommendations 32:35 Is Optimization the Right Language to Understand Deep Learning? - Sanjeev Arora 1:07:19 The Epistemology of Deep Learning - Yann LeCun 59:00 An Introduction to Graph Neural Networks: Models and Applications 43:04 Deep Networks Are Kernel Machines (Paper Explained) 1:51:35 NeurIPS 2020 Tutorial: Deep Implicit Layers 1:14:52 Lecture 7 - Deep Learning Foundations: Neural Tangent Kernels 27:53 Representational Power of Graph Neural Networks - Stefanie Jegelka 02:02 Thirty years of proof: an interview with Andrew Wiles on the anniversary of Fermat's Last Theorem 1:24:44 Stanford Seminar - Information Theory of Deep Learning, Naftali Tishby 1:37:30 Kernels! 51:06 Intro to graph neural networks (ML Tech Talks) 1:17:05 From Deep Learning of Disentangled Representations to Higher-level Cognition 1:16:53 Graph Representation Learning (Stanford university) 38:27 ICLR 2021 Keynote - "Geometric Deep Learning: The Erlangen Programme of ML" - M Bronstein 14:08 The Incredible Potential of Superconductors 20:44 Non-uniqueness Phenomena in Fluid Mechanics - Stan Palasek 1:23:49 Graph Nets: The Next Generation - Max Welling Similar videos 05:22 2 2 3 Neural Tangent Kernel 1:16:51 Lecture 2: The Wide limit of Neural Networks: NNGP and NTK (English) 04:28 Neural Tangent Kernels 1:06:16 Deep Learning Foundations: Simon Du's Talk on Passive and Active Multi-Task Representation Learning 41:40 Day 3 | DAI 2021 | Simon Du 14:09 Learning over-parametrized neural networks-Going beyond NTKs More results