/explore

Click through on any links that interest you or select the planets on the right to continue exploring the Outer Web.
You are here

 dynomight.net
First-principles on AI scaling
| | blog.moonglow.ai
Three Kuhnian Revolutions in ML Training
2.6 parsecs away

Travel
| | Parameters and data. These are the two ingredients of training ML models. The total amount of computation ("compute") you need to do to train a model is proportional to the number of parameters multiplied by the amount of data (measured in "tokens"). Four years ago, it was well-known that if
| | www.alignmentforum.org
New Scaling Laws for Large Language Models - AI Alignment Forum
2.4 parsecs away

Travel
| | On March 29th, DeepMind published a paper, "Training Compute-Optimal Large Language Models", that shows that essentially everyone -- OpenAI, DeepMind...
| | windowsontheory.org
Injecting some numbers into the AGI debate - Windows On Theory
2.4 parsecs away

Travel
| | [Yet another "philosophizing" post, but one with some actual numbers. See also this follow up. --Boaz] Recently there have been many debates on "artificial general intelligence" (AGI) and whether or not we are close to achieving it by scaling up our current AI systems. In this post, I'd like to make this debate a bit...
| | journal.otessa.org
* Call for Papers: Special Issue on Mapping Terminology, Theory, Policy, and Resources Across Key Areas of Impact In Educational Technology | The Open/Technology in Education, Society, and Scholarship Association Journal
16.1 parsecs away

Travel
| [AI summary] The OTESSA Journal is soliciting short papers regarding terminology, policy, theory, and resources for educational technology, with a specific focus on the integration of artificial intelligence.