ml

Asianometry explains the EUV light source inside ASML's lithography machines, where lasers hit falling tin droplets twice to produce 13.5 nm light. It walks through the engineering of the droplet generator, the pre-pulse and main laser, and the plasma that results.

Branch Education's animated deep dive into ASML's EUV photolithography machine and how it prints nanometer-scale transistors onto chips. It covers the system's components and the process of chip manufacturing.

Veritasium explains how ASML's EUV lithography machines make the most advanced computer chips, covering the engineering of the light source, mirrors and the supply chain around them. It is a long-form explainer on the world's most complex commercial machine.

Paul Irish explains how pointerpointer.com works. It uses a hidden canvas of colored Voronoi regions to look up which photo matches the cursor position.

Wes Bos breaks down the techniques behind the McMaster-Carr website's speed: server-rendered HTML, prefetching, CDN and service worker caching, preloading, critical CSS and fixed-size images. He also notes its old-school jQuery and YUI JavaScript.

A Veritasium video on AlphaFold, DeepMind's protein structure prediction system, and the protein design work it enabled. It features interviews with John Jumper and with David Baker's protein design group and explains why predicting protein folding matters.

Michael Wesch's five-minute video about Web 2.0, showing how the read/write web, XML and user-generated content change how people create and share information.

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.

3Blue1Brown's visual explanation of how transformers and GPT-style language models work, tracing data through tokens, embeddings and attention.

Veritasium explains why analog computing may suit AI workloads, covering old analog computers, The Analog Thing and Mythic's analog chips that use flash memory for neural network inference. It argues digital hardware is hitting limits.

A 36-minute animated deep dive from Branch Education into how DRAM works. It shows how memory connects to the CPU and how billions of bits are stored in nanoscale capacitor cells. It also covers why DRAM sits between SSDs and the processor.

3Blue1Brown's visual walkthrough of the attention mechanism inside transformers, covering self-attention, multiple heads and cross-attention. It is chapter 6 of his deep learning series.

Yannic Kilcher walks through the Attention Is All You Need paper, which introduced the Transformer, a sequence model based solely on attention without recurrence or convolutions. He reviews the architecture and the paper's translation results.

Ben Eater gives a detailed look at how an NPN bipolar junction transistor works and what it does. It is an eleven-minute explanation aimed at building real understanding of the device.

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.

Robert Miles explains reward modelling, in which an AI learns what humans want from their preference comparisons instead of a hand-written reward function. He walks through the 2017 paper on deep reinforcement learning from human preferences.

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

Jen Simmons explains the difference between CSS Flexbox and CSS Grid, and when to use each. She makes the case that Grid does not replace Flexbox and most sites use both.

LearnCode.academy presents a roadmap of web development technologies to learn in 2016 and 2017, spanning front-end, back-end and DevOps tools.

Aaron Francis walks through how PHP has changed from 2012 to 2023, covering traits, array destructuring, generators, typed properties and more. It is a quick chaptered tour of modern PHP features.

Rob Miles explains on Computerphile how language models generate plausible text and how transformers improved on recurrent networks, with a look at what comes next.

A short cartoon explaining how language models like ChatGPT work, from autocomplete and conditional probabilities to neural networks, gradient descent and backpropagation. It shows how networks act as universal function approximators.

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.

A Welch Labs explainer on neural scaling laws and the compute-efficient frontier, asking why model loss seems to hit a line it cannot cross. It draws an analogy to an ideal gas law for AI.

Two Minute Papers covers OpenAI's multi-agent hide-and-seek experiment, in which agents learned emergent tool use and found ways to break the game's rules. It summarizes the 'Emergent Tool Use from Multi-Agent Interaction' paper.

Asianometry explains why Russia cannot replace TSMC after sanctions halted chip shipments. It looks at Russian chip designers such as Baikal, MCST and Yadro, and at Russia's lagging domestic manufacturing capacity.

Matt Parker of Stand-up Maths demonstrates MENACE, Donald Michie's 1961 machine made of matchboxes that learns to play noughts and crosses through reinforcement. He builds and trains a physical version.

Asianometry explains that the worst of the 2021-22 chip shortage sits in trailing-edge semiconductors, not the leading-edge chips usually covered. It covers why those older-node fabs are so hard to expand and fix.

A Two Minute Papers episode on NVIDIA's recurrent denoising autoencoder, which cleans up noisy Monte Carlo path-traced image sequences. It shows how a neural network can produce usable renders from very few samples.

Asianometry's short analysis of SMIC's N+2 node, China's 7nm leap forward. It considers how far behind the competition SMIC still is.

Branch Education walks through a semiconductor fab and the dozens of steps used to turn silicon wafers into microchips, with 3D animation. It covers lithography, deposition and etching at a high level.

Asianometry explains why India, a chip design powerhouse, cannot fabricate chips, tracing the 1980s government attempt to build a national semiconductor champion and how it collapsed. It includes the sabotage at the SCL fab.

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.

Yosh trains a reinforcement learning agent to drive Trackmania until he can no longer beat it. The video explains how the training works and the design choices involved.

Tom 7 (suckerpinch) delivers a long, comedic digression on gradient descent and floating-point imprecision. He shows how rounding behavior can supply the nonlinearity needed to build a neural network.

Tom Scott's video whose title shows its own live view count, and the story of how he made that work. It explains why open APIs were once common and why they largely disappeared.

2swap explores lambda calculus using John Tromp's lambda diagrams, with animated beta-reductions, asking what PLUS times PLUS gives. It touches on Church encodings and the equivalence of Turing machines and lambda calculus.

3kliksphilip summarizes Valve's GDC talk on VACNet, a deep-learning system that flags CS:GO cheaters and feeds the Overwatch review queue. It covers Trust Factor, VAC, and how the data is collected and analyzed.

Asianometry traces the 30-year rise and peak of Japan's semiconductor industry from the 1950s to the 1980s, covering how a postwar country gained competence in an emerging technology.

Two Minute Papers covers a paper on speeding up fluid and smoke simulation with convolutional networks. It contrasts the approach with earlier regression forest methods.