deep learning

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.

Mark Rober and Jabril build an app that uses machine learning to decode baseball sign sequences. The video also explains neural networks in a simple, accessible way.

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.

SethBling shows MarI/O, a program using neural networks and the NEAT genetic algorithm to learn to play Super Mario World. The description links the source code, the paper and background reading.

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 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.

OpenAI's short demo of agents playing hide-and-seek in a simulated environment. Through self-play, the agents develop six progressively more complex strategies and counter-strategies, including tool use.

A short video accompanying the CVPR 2018 paper Learning to See in the Dark. It shows a neural network producing sharp images from very low-light raw sensor data, compared against denoising approaches.

Deepfake artist Shamook's improved version of the Luke Skywalker scene from The Mandalorian season 2 finale, replacing the official CGI face with a deepfake.

A demo of PREDATOR (TLD), a real-time object tracker that learns the appearance of its target and improves while it tracks. The technology was later developed commercially.

Toyota Research Institute presents an approach for teaching robots new dexterous behaviors from human demonstration. It explains why this capability matters and what advances made it possible.

Microsoft Research's Rick Rashid demonstrates a speech recognition and translation system, built on deep neural networks, that turns his spoken English into Chinese speech in 2012.

A short Unitree promo for its G1 humanoid robot, priced from $16K, showing agile motion, dexterous hands and imitation and reinforcement learning.

Tom7 explains and demos software that learns to play NES games by finding score patterns in memory and then searching for inputs that raise it. It is the video companion to his SIGBOVIK paper.

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.

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.

A 1970 demonstration of W. Timothy Gallwey, author of The Inner Game of Tennis, teaching a complete beginner to play tennis in about ten minutes. He chats with her, feeds balls and gives positive feedback instead of instruction.

A five-minute SuperMemo demo of incremental reading: importing a Wikipedia article on nutrition, extracting passages, making cloze deletions and finishing the item. It lists the keyboard shortcuts used.

Destin of Smarter Every Day learns to ride a bicycle whose steering is reversed, taking about eight months, and compares it with how his young son learned much faster. It illustrates knowledge versus understanding and brain plasticity in adults versus children.

A bot built by Robin Baumgarten plays Infinite Mario using A* path-finding, shown as a red planned path that updates as new obstacles appear. It won the Mario AI competitions and uses only on-screen information.

Smarter Every Day host Destin Sandlin tries to hover a helicopter and finds it very hard. The video shows how the constant coordinated corrections across several control axes feel at first and how learning makes them automatic.

Sentdex demonstrates GAN Theft Auto, a GameGAN-based neural network that recreates GTA V so the network itself is the playable environment. He shows the results and releases the code.

Karl Sims's 1994 demonstration of evolved virtual creatures: block-bodied creatures whose bodies and controllers are evolved through selection, mutation and recombination to swim, walk and compete.

A video presenting an Intel research paper on enhancing the photorealism of synthetic images. A convolutional network uses rendering intermediates, shown on GTA V footage that comes out looking like real street scenes.

A restored 1994 video of Karl Sims' research on evolved virtual creatures, where simulated block creatures are bred through artificial evolution to swim, walk, follow and compete. It is a landmark demonstration of genetic algorithms in physics simulation.

A short demo video for NVIDIA's StyleGAN paper, which proposes a style-based generator for GANs. It shows face synthesis with style mixing and scale-specific control over attributes like pose and freckles.

Mike Boyd takes up Smarter Every Day's challenge to learn to ride the Backwards Brain Bike, a bicycle whose steering is inverted, and rides 50 meters. It shows how hard it is to unlearn a motor skill.

GeoGuessr pro Rainbolt plays head-to-head against an AI built by Stanford students that guesses locations from street-view images. The video shows the AI repeatedly beating him.

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.

The CVPR 2016 oral demo of Face2Face, a system that captures a person's face and reenacts another video's face in real time using only RGB video. The authors frame it as research on modern vision and graphics.

Technology Connections gives first impressions of using an Onyx Boox Max2 e-ink tablet for productivity, including using it as an external display. He shows what the device can and cannot do.

A short overview of Tabs Outliner, a Chrome extension that shows browser tabs and windows as a nestable tree. It demonstrates organizing, saving and restoring tabs.

Drummer Larnell Lewis hears Metallica's Enter Sandman for the first time on Drumeo and learns to play it by ear, using his musical vocabulary.

A short Tesla clip showing the Optimus humanoid robot sorting objects autonomously and doing yoga-style stretches. Tesla says its neural network is trained end to end, from video in to controls out.

Tom7 (suckerpinch) builds and evaluates a series of absurd hard drives, including storage made from ping packets across the internet. The video accompanies his paper and mixes deep technical hacking with comedic storytelling.

SethBling uses Super Mario World glitches to inject and run Flappy Bird code on an unmodified SNES, entering it purely with controller inputs. It was the first time a human completed this kind of exploit by hand.

A live mock programmer interview in which Casey Muratori interviews game developer Shawn McGrath using a less confrontational format devised by Chris Hecker. It gets deep into technical details while showing how the candidate actually thinks and works.

Mike of mikeselectricstuff shows his heavily modified old BBC Micro system in a 28-minute tour of 8-bit hardware archaeology.

Musician Thoj performs a live looping session on the Teenage Engineering OP-1 synthesizer, building a deep house track layer by layer.

Cosmo's 22:38 Ocarina of Time speedrun, performed live at AGDQ 2013. The runner uses a complex exploit to skip much of the game.