kokopelli

We Only Learn from Error

Posted by kokopelli on Saturday April 11, 2026 @12:05AM
from the nathanclonts.com dept.

https://nathanclonts.com/we-only-learn-from-error/

( Read More... | 1 comment | on HN )

Re:We Only Learn from Error on Saturday April 11, 2026 @12:05AM (#47727285)

I've been learning robotics by training a vision-language-action model, deployed to a Raspberry Pi with a pan/tilt camera mount, to track faces based on text commands. This article covers one of the issues I ran into with creating a useful simulation environment.

Re:Starfling: A one-tap endless orbital slingshot game in a single HTM... on Wednesday April 08, 2026 @10:49PM (#47698777)

This is great. Very addicting.