deep modules

Max Tegmark's MIT CBMM talk on why deep learning works so well. He argues that the laws of physics and the symmetries in physical data make it compressible into small neural networks, tying physics to network structure.

A short smartphone-recorded clip from a 2014 event in Paris where DeepMind CEO Demis Hassabis presents an AI learning to play old Atari games. The system learns from only the video signal and the score. Audio quality is poor.

Geoff Hinton's 2010 Google Tech Talk on deep belief networks, learning layers of representation one at a time from unlabeled data, and extending them with multiplicative interactions.

A deepfake in which Bill Hader, impersonating Arnold Schwarzenegger on a talk show, visibly morphs into Schwarzenegger. It was made with DeepFaceLab by Ctrl Shift Face.

A Mandelbrot set zoom animation reaching a magnification of about 2.1x10^275, claimed as the deepest at the time (January 2010). Each frame was rendered separately over six months on 12 CPU cores.

A six-minute journey through the layers of a neural network using Deep Dream, with each frame fed back into the network starting from random noise and the targeted layer changing every four seconds. It is set to downtempo music.

John Ousterhout, Stanford professor and creator of Tcl, presents the ideas of his book A Philosophy of Software Design in a Talks at Google session. He covers managing complexity, deep modules and interfaces, and what he teaches in his Stanford software design course.

Martin Gorner's Devoxx talk introduces deep learning with TensorFlow, walking through neural networks for MNIST, how to choose an architecture and how to diagnose networks that fail to train.

Andrew Ng presents unsupervised feature learning and deep learning, showing how algorithms can learn representations from unlabeled data instead of hand-designed features. He gives examples in vision and audio, including visualizations of features learned by successive layers.

The first lecture of Stanford's Natural Language Processing with Deep Learning course, taught by Chris Manning and Richard Socher. It introduces NLP and word vectors, including SVD, skip-gram, CBOW, negative sampling and word2vec.

A two-minute Insider Tech clip on DeepMind's simulated agents that learned to walk, run, jump and climb without guidance, with the goofy resulting gaits.

3Blue1Brown's first chapter on deep learning explains what neurons and layers are and the math underlying a basic neural network, using handwritten digit recognition with visual animations.

A GDC 2018 talk by Valve's John McDonald on VACNet, a deep learning system that analyzes CS:GO gameplay to detect cheaters. It covers how the model is trained on human-reviewed Overwatch cases and how its output feeds human moderation.

Andrew, diagnosed with early onset Parkinson's at 35, shows how deep brain stimulation surgery has controlled his motor symptoms. The short video shows his tremors with the stimulator off and on.

3Blue1Brown's third deep learning chapter gives a visual, intuitive account of backpropagation and what happens to a neural network as it learns. A follow-up video covers the underlying calculus.

ArtosisTV's commentary on Game 1 of DeepMind's AlphaStar (Protoss) against top Zerg player Serral at Blizzcon 2019. Artosis analyzes the match with beginner-friendly commentary.

A DeepSec 2010 conference talk by Ralf Philipp Weinmann on attacking the baseband processor of mobile phones rather than the application processor. It covers how open-source GSM base stations make malicious-tower attacks practical against neglected GSM stacks.

A BuzzFeed PSA from 2018 in which a deepfaked Barack Obama, voiced by Jordan Peele, warns about fake video before revealing the trick. It is a one-minute demonstration of AI face and voice manipulation.

3Blue1Brown's second neural network chapter explains cost functions and gradient descent, showing how a network learns by minimizing error during training.

Part two of EEVblog's visit to the NASA/CSIRO Canberra Deep Space Communication Complex, showing the control room and how operators contact Voyager 2. It includes live data and spectrum displays.

Dave Jones of EEVblog and Richard Stephenson tour the 70m DSS43 dish at the Canberra Deep Space Communication Complex and explain how NASA talks to Voyager 2. It covers the transmitters, receivers and tracking involved.

Rick Houlihan's AWS re:Invent 2018 deep dive on advanced DynamoDB design patterns. It covers index overloading, GSI sharding, composite keys, graph processing and transactions.

Sacha Arnoud, Waymo's Director of Engineering, gives a lecture in MIT's 2018 deep learning for self-driving cars course on the rise of machine learning in autonomous vehicles. It covers how Waymo uses deep learning and how it handles tricky driving situations.

3Blue1Brown's chapter 4 on deep learning gives a more formal, symbol-based account of backpropagation calculus, connecting the intuition from the previous chapter to the math found in textbooks and code.

Sean Carroll's Talks at Google presentation of his book Something Deeply Hidden. He gives an accessible introduction to quantum mechanics and argues for the many-worlds interpretation and the emergence of spacetime.

A short demo of the Deep Visualization Toolbox, which shows the activations inside a convolutional neural network as it processes live camera input. Viewers see individual units responding to edges, faces and text.

Engineer Sandy Munro joins an Autoline panel for a 100-minute deep dive on Tesla's design and manufacturing, including the Model 3 and its motors.

Inventor Mary Lou Jepsen demonstrates using red light and sound to see inside the body and brain, aiming at tumor detection and neural activity measurement, and potentially a cheaper wearable MRI replacement.

Demis Hassabis of DeepMind speaks at Google Zeitgeist, saying human-level AI is decades away but the debate should start now. He covers DeepMind's focus on game-playing agents.

A BBC News segment on University of Washington research that synthesized a photorealistic Obama video, modeling mouth movement so any audio can be put in his mouth.

The first lecture of David Silver's reinforcement learning course, introducing the field and its basic concepts. The course follows the Sutton and Barto textbook.

A 29-minute cinematic timelapse by melodysheep that journeys trillions of years into the future, covering the death of the sun, the end of stars, proton decay, black hole evaporation and the fate of the cosmos.

Rich Hickey's philosophical talk from the first Clojure Conj in 2010. He argues that good design comes from deliberate, away-from-keyboard thinking, and describes how to feed the background mind so it solves problems.

Adam Conover's talk at a millennial marketing conference argues that generations are an unscientific, condescending idea and that millennials do not exist as a real category. The description notes that researcher Jean Twenge later said he misrepresented some of her work.

Ali Rahimi's NIPS 2017 Test-of-Time award talk, in which he argues that modern machine learning has become alchemy, working without being rigorously understood. He calls for more rigor and understanding in the field.

agadmator walks through a game in which DeepMind's AlphaZero beat Stockfish, ending in a famous zugzwang that leaves the engine helpless. He gives commentary on the moves.

Adobe's VoCo demo from the MAX 2016 Sneak Peeks, co-hosted by Jordan Peele, showing software that changes words in a voiceover by typing new ones.

A Microsoft Build 2019 session in which the WSL team explains the new WSL 2 architecture, which runs a real Linux kernel alongside Windows on a hypervisor. It covers faster file performance, Linux file access from Windows and better application support.

The first lecture of Stanford's CS231n course on convolutional neural networks for visual recognition. It surveys the history and key challenges of computer vision, from the Cambrian explosion and Hubel and Wiesel to ImageNet.

Mobileye co-founder and CTO Amnon Shashua explains what goes into sensing for autonomous driving and whether end-to-end deep learning can work for it. He argues for decomposing the problem.