bayesian inference

Michael Betancourt's lecture at MLSS Iceland 2014 on efficient Bayesian inference with Hamiltonian Monte Carlo. It covers MCMC concepts and motivation and the problems of sampling in high-dimensional distributions.

Veritasium explains Bayes' theorem with examples and introduces the 'Bayesian trap', where repeated events get mistaken for inevitable ones. It discusses how priors shape our expectations and behavior.

Julia Galef uses pictures to explain Bayes' rule, then tells three personal stories showing how she applies it to improve her thinking.

A four-minute minutephysics video arguing that causal models and causal networks let statistics infer causation from correlation. It touches on Reichenbach's principle, common causes, feedback and quantum causal discovery.

Neuroscientist Karl Friston gives a 15-minute talk on the free energy principle, covering Markov blankets, Bayesian model evidence and global brain theories.

A doctor from Medlife Crisis explains in under three minutes why even a rich patient should not simply order every medical test, touching on pre-test probability and Bayesian reasoning. False positives make blanket testing counterproductive.

Peter Norvig's 2007 Google talk on theorizing from data, covering what huge datasets can do for language understanding, translation, information extraction and inference. It argues that more data often beats cleverer models.

Guy Steele's PPoPP 2017 keynote surveying the history of computer science metanotation, including BNF, inference rules, and overline notation. He argues the field's notations are inconsistent and fragmented.

The first lecture of David MacKay's Cambridge course on information theory, pattern recognition and neural networks. It introduces the core ideas of his book Information Theory, Inference, and Learning Algorithms.

Julien Verlaguet of Facebook describes applying static analysis and type inference to a very large PHP codebase, using OCaml. The work led to the Hack language.

SciPy 2015 talk by Nathaniel Smith and Stefan van der Walt on choosing a new default colormap for Matplotlib. It covers why Jet misleads, how perceptual color spaces work, and how viridis was designed.

Hans Rosling uses data and animated statistics to show how world population is changing, why birth rates fall as child survival and wealth rise, and why population is expected to peak around 11 billion. It also tackles common misconceptions about Africa and family planning.

Veritasium explains why a large share of published research findings may be false, covering p-values, p-hacking, publication bias and low replication rates. It draws on John Ioannidis's well-known paper and describes reforms such as preregistration.

Hans Rosling and his son Ola run an audience quiz with global population, health and income data to show how wrong most people are about the world. Ola then gives four ways to get less ignorant.

Joel Grus's JupyterCon talk explaining why he dislikes Jupyter notebooks. He covers hidden state, trouble with reproducibility and testing, and how notebooks frustrate teaching. He shows his preferred workflow and what Jupyter could change.

A five-minute lightning demo of VisiData, a terminal tool for exploring and manipulating data. It uses two months of NYC 311 complaint data to drill down, plot points on a virtual map and save selected rows.

John Kitchin shows how Emacs org-mode with Python lets researchers combine narrative, math, figures and executable code in one document. It exports to LaTeX, HTML and PDF, and he used it for papers and e-books.

Jeremy Howard defends Jupyter notebooks as a serious development environment, especially when paired with nbdev. He responds directly to Joel Grus's talk about why he dislikes notebooks, and contrasts REPL-style and notebook-style work.

A reupload of Joel Spolsky's talk teaching spreadsheet users the Excel features they are probably missing, such as keyboard navigation, named ranges, tables and lookups. The original disappeared and this copy comes from the Wayback Machine.

Database pioneer Michael Stonebraker argues that 'big data' is really at least four different problems: big volumes with small analytics, big analytics, velocity, and variety. He comments on where each market is heading.

Fons van der Plas introduces Pluto.jl, a reactive notebook for Julia that tracks references between cells and re-runs affected cells when code changes. The talk was given at JuliaCon 2020.

The first video in a 55-part series by Michel van Biezen that introduces the Kalman filter and builds from the rationale through algebraic examples to the full matrix treatment. The videos are short, blackboard-style lectures.

Dan Robinson, then CTO of Heap, explains how Heap runs its product analytics backend on PostgreSQL and CitusDB. It covers conversion funnels, retention and cohort queries across tens of billions of events and 50 TB of data, plus Postgres and Kafka together and the indexing strategy.

Ezra Klein interviews Democratic pollster David Shor, who walks through survey data and charts explaining the 2024 election results. They discuss voter shifts, the education divide and what Democrats got wrong.

David Beazley's PyData Global 2021 keynote, an oral history of early scientific Python. He tells how Python ended up running on a Connection Machine supercomputer at Los Alamos.

A clip from Yes Prime Minister in which Sir Humphrey shows how leading questions can make a survey support or oppose national service. It is a short satirical scene.

In this TED talk, Hans Rosling uses animated statistics to debunk myths about the developing world, showing how health and income have improved across countries.

David Kriesel's 36C3 talk, in an English translation, analyzes a year of scraped Deutsche Bahn data covering every stop of every long-distance train in Germany. It explores delays and how punctuality is actually measured.

Sandy Pentland of MIT presents his book on social physics, which uses big-data studies of how people interact to explain how ideas spread. He describes findings on collective intelligence and social influence.

Robert Erdmann, a materials scientist at the Rijksmuseum, gives a PyCon keynote on the ultra-high-resolution imaging of Rembrandt's Night Watch. He covers the automated gantry camera system, the Python software that drives it, and the large volumes of image data involved.

A short Veritasium video explaining regression to the mean and how ignoring it leads to wrong conclusions about punishment versus reward, medical treatments and sports streaks. It uses the example of flight instructors who believed criticism works better than praise.

John Rauser's Strata 2014 keynote argues that programmers can learn statistics quickly by simulating random processes and shuffling data. He explains p-values through resampling rather than formulas.

Mark Raasveldt of CWI presents DuckDB, an embeddable analytical database described as the SQLite for analytics, in a CMU Database Group talk. He explains why it was built, what it is for, and how it works.

Coffee Break analyzes about 40,000 videos from seven months of YouTube's trending tab. It examines whether trending is biased toward mainstream media and late-night shows.

Lex Fridman interviews Travis Oliphant, creator of NumPy, SciPy and Anaconda, about scientific Python and its history. It runs about three hours.

Alexander Nix of Cambridge Analytica presents at the 2016 Concordia Summit on using big data, psychographic profiling and behavioral microtargeting in elections.

Peter Norvig's UBC distinguished lecture on how massive amounts of simple data, such as for language modeling, can yield understanding that hand-built models could not.

Veritasium explains why many real-world systems follow power laws rather than normal distributions. It covers self-organized criticality, with simulations of sandpiles, magnetism and forest fires.

Jan Vitek of Northeastern University gives a meetup talk on the challenges of making the R language run fast. He explains what in R's semantics makes it hard to optimize and to build a JIT compiler for.

Jake Vanderplas's PyCon 2016 talk shows how sampling-based methods such as simulation, shuffling and bootstrapping make statistics intuitive. His pitch is that if you can write a for-loop, you can do statistics.