self-paced learning

Dwarkesh Patel interviews Andy Matuschak on how he learns, why memorization matters for understanding, and why most learning tools fail. The conversation also covers handling confusion and the role of tutoring.

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.

Machine Learning Street Talk interviews Paul Lessard about Categorical Deep Learning, which uses category theory and monads to make neural network architectures more interpretable and composable. They also discuss neural networks' limits in reasoning and unbounded computation.

The first lecture of Andrew Ng's Stanford CS229 Machine Learning course, giving an overview of the course. It introduces supervised and unsupervised learning, learning theory and reinforcement learning.

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.

Chris Urmson's SXSW 2016 talk on Google's self-driving car project, covering 1.3 million self-driven miles and the gap between hype and reality. It shows sensor visualizations and playbacks of hard situations the cars met, and discusses failures such as the low-speed bus collision.

Valve's John McDonald presents at GDC 2018 on how Valve uses deep learning (VACnet) to detect cheaters in Counter-Strike: Global Offensive. He covers training on human-reviewed replays and how model output is confirmed by manual review.

The first episode of Google's Machine Learning Recipes series. It introduces machine learning and builds a first supervised-learning classifier in about six lines of Python using scikit-learn.

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.

David Ungar of IBM Research reflects at Stanford on the Self language project and argues the real legacy of ideas is the people. He covers its history, philosophy, successes and failures.

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.

George Hotz of Comma AI gives a 2017 conference talk on lessons from building a self-driving system. He argues about how driving should be defined and why the common approach to autonomous vehicles is flawed.

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 1995 Sun Microsystems film about the Self programming language, a prototype-based Smalltalk relative, showing its live, graphical programming environment.

Sal Khan of Khan Academy argues that schools push students forward with gaps in their foundations, and that students should instead master concepts at their own pace. He compares it to building a house on an unfinished foundation.

A clip from Tesla Autonomy Day 2019 where Elon Musk argues cameras are enough for self-driving and calls lidar a fool's errand.

AI Addict reviews a Tesla Full Self-Driving Beta drive, showing hard braking, red-light running, wrong-way turns and a collision with a bollard. The video is timestamped by incident.

Ali Rahimi accepts the NIPS 2017 Test of Time Award and argues that machine learning has become alchemy, with practice running ahead of understanding.

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.

Andrej Karpathy, then Tesla's head of AI, presents at ScaledML 2020 on how Tesla builds full self-driving using vision-based neural nets and fleet data collection. He covers training custom detectors and neural nets gradually replacing traditional code.

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.

The first lecture of Caltech's CS 156 Machine Learning course by Yaser Abu-Mostafa, introducing the learning problem and supervised, unsupervised and reinforcement learning. It is one of 18 lectures in the Learning from Data course.

Lex Fridman interviews George Hotz, founder of Comma.ai, in a three-hour conversation. Topics include self-driving, Tesla and Waymo, crypto, GPT-3, immortality and book recommendations.

The introductory lecture of Richard Hamming's capstone course, The Art of Doing Science and Engineering: Learning to Learn. It focuses on style of thinking and on educating rather than training, using technical material only as illustration.

Part 2 of the IROS 2011 keynote by Sebastian Thrun and Chris Urmson on Google's self-driving car. It shows how the car perceives its surroundings with a lidar point cloud and recognizes traffic lights and pedestrians. The audio is poor because it was recorded on a pocket camcorder.

Mark Rober tests whether self-driving cars can be fooled, comparing a lidar-equipped car with a camera-only Tesla in scenarios including a painted Wile E. Coyote wall and smoke.

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 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.

The final lecture of Richard Hamming's Learning to Learn course, from 1995, based on his well-known talk 'You and Your Research'. It covers how to do important work and build a career.

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.

Not Just Bikes argues that self-driving cars will worsen sprawl and car dependence and harm cities. The video compares car-centered and people-centered urban design and weighs what cities should do instead.

3Blue1Brown explains fractal dimension and shows why many fractals are not strictly self-similar. Self-similarity turns out to be a special case of the broader idea of fractional dimension.

Veritasium's Derek Muller gives a Perimeter Institute livestream on how AI might change teaching and learning. He draws on past technology shifts and cognitive science to argue that learning requires effortful repetition.

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.

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.

Lex Fridman's long conversation with Andrej Karpathy on Tesla's self-driving approach, Optimus, neural networks, transformers, AGI and more. It also touches on his OpenAI work and working with Elon Musk.

Lesson 1 of fast.ai's Practical Deep Learning for Coders, taught by Jeremy Howard. It takes coders from running code in a Jupyter notebook straight into training a real image model.

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.

Yandex shows its self-driving taxi navigating snowy Moscow streets after a heavy snowfall, dealing with traffic, pedestrians, parked cars and snowbanks on real roads.

Ben Krasnow of Applied Science shows that a plain laser diode can measure sub-micron vibrations from centimeters away using self-mixing interferometry, and can also be used for range-finding.