learning science

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.

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.

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.

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.

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.

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.

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.

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.

A Berkeley Lab Distinguished Lecturer talk on building science, with an expert covering control layers, insulation and vapour barrier placement, and how buildings fail. The delivery is humorous.

A short clip of Neil deGrasse Tyson answering a question about why there are not more women in science. He describes the resistance he faced as a Black kid pursuing science and the systemic barriers many people encounter.

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.

An excerpt from Hal Abelson's opening SICP lecture introducing computer science and contrasting declarative and imperative knowledge. He argues that computer science is not really a science and is more like magic or engineering of ideas.

A two-minute clip of Richard Feynman on the social sciences, discussing what counts as real science versus pseudo science.

A John Oliver Last Week Tonight segment on how forensic science used in criminal trials is often unscientific. It looks at unreliable methods and lab malpractice, and ends with a joke about a TV procedural that might change public perception.

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

A Capella Science parody of Ed Sheeran's Shape of You, rewritten about molecular bonding and orbitals and sung a cappella.

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.

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.

Physicist Angela Collier's roughly hour-long video on how string theory was hyped to the public and what that did to science communication. She traces the history of the theory as a narrative and argues it is interesting but was oversold.

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.

Andrei Tarkovsky's 1972 film Solaris, in which a psychologist is sent to a space station orbiting a planet with an alien intelligence and meets his dead wife there. It is the full movie, uploaded by Mosfilm.

Freeman Dyson's 2005 Boston University talk on heretical views about science and society. He questions prevailing dogmas, including his argument that climate and CO2 are a land management issue.

A Symphony of Science music video that auto-tunes samples of Carl Sagan, Richard Feynman, Neil deGrasse Tyson and Bill Nye into a song about the wonder of the universe.

A Capella Science parody of Despacito that sings through evolutionary developmental biology, explaining how a single cell becomes a body, including body-axis genes and dormant snake leg DNA. It runs about four minutes.

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.

A two-minute clip of Richard Feynman explaining why he considers much of social science pseudoscience. He stresses how easy it is to fool yourself and how much care real knowledge takes.

First lecture of MIT 6.00, an introductory computer science course taught in Python by Eric Grimson and John Guttag. It covers the goals of the course, what computation is, and data types, operators and variables.

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.

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.

Andrew Huberman introduces how the nervous system produces sensation, emotion and behavior, and how neuroplasticity lets us change it. It covers neurons, synapses, brain chemicals and the rhythms governing focus, learning and sleep.

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 clip from It's Always Sunny in Philadelphia season 8 in which a character argues that science is a liar and cannot be trusted.

A Capella Science parody of Camila Cabello's Havana, retold as a song about molecular machines and nanotechnology. It features animations of rotaxanes, catenanes and molecular switches, and comes with footnotes.

A two-minute 1981 interview clip in which Richard Feynman answers an artist friend who claims science takes the beauty out of a flower. Feynman argues that understanding adds to beauty and mystery.

A DEF CON 26 talk by investigative journalists and hackers on the fake science industry, with bogus conferences and predatory journals that give studies false credibility. It draws on a large investigation into the money behind them.

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.