machine learning

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.

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

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.

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.

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.

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

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.

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.

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.

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.

Johnny Harris investigates why McDonald's ice cream machines are so often broken, covering the Taylor machines, the repair restrictions on franchisees and the Kytch workaround device.

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.

Technology Connections takes apart the step-by-step electromechanical logic of a 1970s pinball machine. He shows how relays, steppers and switches track state and score without electronics.

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.

Veritasium debunks the idea of learning styles, such as being a visual learner, with an interview with Daniel Willingham and a memory experiment. It argues that a multimodal approach helps everyone.

Technology Connections takes a long, detailed look at how a 1970s electromechanical pinball machine works, from relays to stepper motors and mechanical logic. It is the first part of a 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.

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.

Robert Miles covers a paper where researchers ran real versions of his earlier mesa-optimizer thought experiments. They show inner misalignment, where an AI learns the wrong goal in training and misbehaves when deployed.

Veritasium short showing Drew Berry's scientifically accurate animations of the molecular machines that drive cell division. It walks through mitosis, chromosomes and motor proteins.

An illustrated 23-minute summary of David Friedman's talk based on The Machinery of Freedom. It describes how law, police and courts could work as market services in a stateless society, covering private law, polycentric law and objections.

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.

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.

Jared Owen's 3D animated breakdown of how a Brunswick GS-X bowling pinsetter works, showing its mechanisms for sweeping, lifting and resetting pins and balls.

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 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.

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.

A guest video by Welch Labs on 3Blue1Brown explaining how text-to-image and video models work. It covers CLIP, shared embedding spaces, diffusion models (DDPM, DDIM), conditioning, guidance and negative prompts.

A Business Insider World Wide Waste episode on H&M's garment recycling machines, which turn used clothing into new apparel. It asks whether a fast fashion company can solve the waste problem it helped create.

Rob Miles discusses GPT-3 on Computerphile, covering how scaling parameters kept improving results over GPT-2 and how the model nearly learned arithmetic without anyone knowing exactly how.

Ray Dalio's 30-minute animated video explains how the economy works through credit, debt, interest rates, and short- and long-term cycles. It is based on his template for understanding the economy.

Veritasium examines which experts have real expertise, drawing on chess research and work by Klein and Kahneman. It explains that expertise comes from pattern recognition built through deliberate practice in valid, feedback-rich environments.

Episode 4 of Mike Maloney's Hidden Secrets of Money, explaining how currency is created through a chain of IOUs among the Treasury, banks and the Federal Reserve. It covers bonds, Fed checks, debt, taxes and fractional reserve lending, from a gold-and-silver advocate's viewpoint.

A short Al Jazeera animation narrated by Amy Goodman presenting Noam Chomsky's five filters of the mass media: ownership, advertising, media elite sourcing, flak and a common enemy. It is part of the MediaTheorised series.

Veritasium explores compliant mechanisms, devices that move by flexing rather than using hinges and joints, with researchers at BYU. It covers advantages such as fewer parts and precision, including uses in nuclear weapon safety.

Veritasium examines research suggesting science explainer videos like Khan Academy's make students feel they understand while tests show they have not learned, because misconceptions go unaddressed. It suggests how to partly fix this.

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.