@pmddomingos
Professor of computer science at UW and author of '2040' and 'The Master Algorithm'. Into machine learning, AI, and anything that makes me curious.
AI
Summarizing Qwen experiments, @gurtej__gill_ says larger, more diverse datasets let supervised fine-tuning rival reinforcement learning.
AI
Anthropic says comparable chip work took dozens of engineers months, versus about three days with Claude.
AI
The Information reports a 50% revenue jump as demand for newer models and enterprise products picked up.
AI
Google’s new model targets demanding work; early testing suggests coding promise and flags a case of scope creep.
AI
Anthropic's planned computing buildout could become harder to finance if Blackstone's forecast of a high junk rating proves right.
AI
The tool builder claims 600-plus bugs and says they asked Claude to cross-check them against merged PRs.
AI
In posts, @yus167 expects simpler infrastructure and says static data and reinforcement learning could mix at any point during training.
AI
One post agrees with the message behind the metaphor but asks whether people think in height, width and depth.
AI
An AI-assisted audit site lists 2,035 alleged problems across 138 papers; Tyler Cowen urged a reassessment.
AI
Ivan Vendrov recalls doubts about Anthropic’s alignment research and a manager’s request that he work on its core API.
AI
Anthropic says its registration statement is not yet effective, so securities cannot be sold or offers to buy accepted.
AI
Developpez says the pledge aims to keep mathematics a rewarding human activity and protect academic careers.
AI
Nassim Taleb predicts sweeping white-collar job losses; another poster jokes about autonomous sprinklers targeting raccoons.
AI
MIT's course page outlines four areas: context, agent harnesses, models and increasingly complex synthetic tasks.
AI
A reply cites the UK and Mistral; the original poster clarifies they meant the EU.
AI
Two posts point to security incidents and urge AI-assisted defenses; one specifically calls for faster vulnerability detection and patching.
AI
Interviewer Matthew Berman lists segments on AI treatments, ChatGPT’s agreeableness and OpenAI’s bet on chips.
XAI
OpenAI credits Calif's model-assisted finding and PHP maintainers' fixes in a response coordinated over two months.
AI
A user shares an image link and jokes that the traces are “totally legible, nothing to worry about.”
AI
The poster says he predicted a 'society of models' in 'The Master Algorithm' in 2015.