financial models

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.

Sasha Rush surveys linear RNNs and state space models as replacements for attention in transformer-style models, in a talk for MLSys. It covers the foundational preliminaries and predates Mamba.

Jeremy Howard's 90-minute practical tour of language models. It covers how they work, GPT-4's capabilities and limits, using the OpenAI API, code writing, data analysis and fine-tuning.

Andrej Karpathy gives a one-hour general-audience introduction to large language models, covering what they are, how they are trained, comparisons to operating systems, and security challenges such as jailbreaks.

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.

A 2006 television appearance in which Peter Schiff warns on Kudlow & Co about the housing bubble and an approaching US economic collapse.

The full 2010 documentary Inside Job, directed by Charles Ferguson, examines the causes of the 2008 global financial crisis. It traces the story through interviews with key financial and political players across several countries.

A VPRO documentary on quantitative analysts who designed the mathematical models and high-speed trading systems behind modern finance. It features figures such as Paul Wilmott and asks what risks come from treating markets as a machine.

A clip from the 2011 film Margin Call in which an analyst explains to the firm's executive the dire state of their investment firm at the onset of the 2008 financial crisis.

A short clip of CNBC's Jim Cramer insisting Bear Stearns is fine in March 2008, days before the firm was sold to JP Morgan for $2 a share.

A Financial Times film examining how Brexit hurt the UK economy, with FT writers and business people. It looks at trade effects, political silence about the costs, and the absence of a convincing Brexit dividend.

A short clip from the film The Gambler in which John Goodman's character gives a monologue about the 'position of fuck you': building enough financial security that you owe nobody anything. The scene is a compact statement of financial independence as freedom.

Yann LeCun on the Big Technology Podcast discusses why current AI models cannot make discoveries despite their knowledge. He argues systems need abstract world models and also covers limits of AI research, investor expectations, and open source after DeepSeek.

Sabine Hossenfelder explains climate sensitivity and why recent observations suggest climate models may have underestimated warming. She says this changed her view on how worried to be.

Former bank regulator William Black explains how fraud, including liar's loans, was committed from inside banks and drove the 2008 financial crisis.

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.

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

Andrej Karpathy's three-hour-plus general-audience deep dive into how ChatGPT-style LLMs are built, covering the full training stack from pretraining to post-training. It also gives mental models for their quirks, like hallucination, and how to use them well.

Dan Olson's long Folding Ideas video on the GameStop meme stock phenomenon, its online community and the conspiracy thinking around it. It runs over two hours.

Audio of Charlie Munger's 1995 Harvard speech on the psychology of human misjudgment, laying out a framework of cognitive biases and mental models for decision making. It covers factors like incentives, social proof and denial.

Martin Fowler's GOTO 2012 introduction to NoSQL databases: where they came from, their data models, how consistency thinking differs, and when to use them. He argues they won't make relational databases obsolete and introduces polyglot persistence.

A clip from the 2011 film Margin Call in which senior partners hold an emergency meeting and decide to sell off toxic assets immediately. It is set at the start of the 2008 financial crisis.

The first part of Extra History's animated miniseries on the South Sea Bubble. It explains how England's government debt led Robert Harley to turn to financier John Blunt, whose schemes began inflating artificial demand for shares.

First session of MIT 15.401 Finance Theory I (Fall 2008) taught by Andrew Lo, giving a course overview and introducing core finance concepts. The course ran during the 2007-08 financial crisis.

A clip from the 2011 film Margin Call in which a firm's CEO, played by Jeremy Irons, tells a senior executive about the cycles of financial crashes. Kevin Spacey's character says he wants out, and the scene ends on the line that it is just money.

Sentdex demonstrates GAN Theft Auto, a GameGAN-based neural network that recreates GTA V so the network itself is the playable environment. He shows the results and releases the code.

A panel with Tony Hoare, Joe Armstrong and Carl Hewitt, hosted by Francesco Cesarini, on the history and future of concurrency. The three creators of CSP, Erlang and the actor model discuss their differing approaches.

A Technology Connections video essay on how analog and digital clocks shape different mental models of time, with the minute hand acting as a progress indicator.

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.

A 2015 interview in which physicist Freeman Dyson, then 91, argues that global warming concerns are exaggerated, that CO2 is beneficial and that climate models oversimplify complex systems. It is posted on a channel that opposes climate consensus.

Jayson E. Street's DEF CON 19 talk shows photos from real security engagements where he was already inside client facilities, illustrating how social engineering and physical intrusion can cause huge losses. For each attack he also explains what defenses would have stopped him.

An independent feature documentary in which 23 economists, advisers and financiers discuss how the financial system works and what a more just economic model might look like. It covers money creation, debt and the aftermath of the financial crisis.

UC Berkeley EECS colloquium by OpenAI's John Schulman on reinforcement learning from human feedback. He covers why RL helps with truthfulness, hallucination, and the limits of behavior cloning.

A one-minute scene from The Gambler in which John Goodman's loan shark explains the 'position of f you': having enough money to walk away from any employer or deal.

Andreas Antonopoulos argues that money has a fourth function beyond store of value, medium of exchange and unit of account, which is as a system of control. He covers the history of money and why it does not flow freely.

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.

Mark Blyth's Talks at Google presentation based on his book on austerity. He argues that government debt came largely from bailing out banks, and that budget cuts are a misguided response to the financial crisis.

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 movie clip from The Gambler (2014) in which John Goodman's character explains to Mark Wahlberg's the idea of having enough money to tell anyone to go to hell. The relevant part is the first 45 seconds.

Andrej Karpathy builds a GPT from scratch in code, following Attention Is All You Need and GPT-2/3, in about two hours using PyTorch. He connects it to ChatGPT and assumes familiarity with his earlier makemore videos.