Comp. sci. prof. @ American University, Washington DC. AI & games researcher with miscellaneous other interests. https://www.kmjn.org/
Mark J. Nelson
Loading...
nice skies in DC as the thunderstorm line cleared out
Die Ausstellung “Captured in Code” zeigt die Vielfältigkeit von Fotografie in digitalen Spielwelten.
11. - 21.6.2026 | Depot Dortmund | Eintritt frei
Eröffnung: Donnerstag, 11. Juni 2026, 19:00 Uhr
Alle Infos, Programm & Anmeldung für die Workshops:
www.capturedincode.de
Now published in open access! Your one-stop shop for the philosophy of language models. It's the spiritual descendant of our two-part preprint from 2024, fully updated. This should be particularly useful for anyone looking for an entry point into this rapidly growing field.
Mark J. Nelson
lol at this report back by an American visiting Tsukuba Expo '85 onlinelibrary.wiley.com/doi/abs/10.1...
I don't always like the practices, but sometimes I do, and even knowing I'm actively rejecting them is helpful. One practice I like is type annotations (which I never used to do). One I don't is extensive error handling (for most code I write, I'd rather let exceptions unwind and crash early).
the U.S. pavilion in question
Intriguing-looking postdoc on human-centric reinforcement learning, part of "an interdisciplinary project between artificial intelligence, social sciences, and societal partners". At TU Eindhoven with Hendrik Baier, deadline July 9. www.tue.nl/en/working-a...
The success of large language models (LLMs) across many domains of AI research has generated intense debate. Some attribute their impressive performance on complex tasks to human-like linguistic and ...
compass.onlinelibrary.wiley.com
An interesting thing about LLMs in Python is that they seem to broadly push code towards some kind of conventional wisdom about best practices, as judged maybe by whoever is setting up the posttraining recipes (I say "in Python" mostly because I notice that more strongly in Python).
Wikipedia tells me this expo is where the Jumbotron was first displayed, as a non-AI note
The most productive scientists are driven by resentment, rivalry, and spite. But too many academics work in collegial obscurity. That’s why I’m offering the opportunity of a lifetime to invest in Tinder for Nemeses
