Accelerated Training by Amplifying Slow Gradients Published 2024-06-20 Download video MP4 360p Recommendations 26:50 Are 1-Bit Weights The Future of Matrix Multiplication?!?!!? 29:47 Grokking: Generalization beyond Overfitting on small algorithmic datasets (Paper Explained) 31:51 MAMBA from Scratch: Neural Nets Better and Faster than Transformers 22:33 Stack Overflow stopped caring about developers a long time ago 40:08 The Most Important Algorithm in Machine Learning 20:18 Why Does Diffusion Work Better than Auto-Regression? 17:38 The moment we stopped understanding AI [AlexNet] 11:14 The Man Who Solved the World’s Hardest Math Problem 35:11 Is the Future of Linear Algebra.. Random? 26:52 The Physics Of Associative Memory 55:55 Miles Cranmer - The Next Great Scientific Theory is Hiding Inside a Neural Network (April 3, 2024) 35:53 Official PyTorch Documentary: Powering the AI Revolution 15:30 C++ Weekly - Ep 435 - Easy GPU Programming With AdaptiveCpp (68x Faster!) 1:03:29 Dertouzos Distinguished Lecture, Prof. Dan Spielman Similar videos 04:26 What is Grokking and Over-Fitting of an LLM 01:42 METEO 300: Friction 09:19 The Evolution of Gradient Descent 2:43:33 Carl Shulman (Pt 1) - Intelligence Explosion, Primate Evolution, Robot Doublings, & Alignment 25:57 Cedar: A Composable ML Accelerated Analog Circuit Simulator | G Hertz, P de Vos | JuliaCon 2021 47:31 Lecture 8 (part 1): Proximal gradient descent, acceleration 26:55 Fast Gradient Boosting Decision Trees with PyGBM and Numba | SciPy 2019 | 2:20:32 Government and society after AGI | Carl Shulman (Part 2) 00:24 Become An Electrical Lineworker 1:38:07 Session 3 44:54 Tero Karras - Training Generative Adversarial Networks with Limited Data More results