"AIXIjs: A Software Demo for General Reinforcement Learning", Aslanides 2017

Comments

sorted by
magical algorithm
Highlighting new comments since Today at 7:21 AM
Select new highlight date
All comments loaded

Source repo: https://github.com/aslanides/aixijs

Live demos: http://aslanides.io/aixijs/demo.html

This is relevant as they can be used to implement AI risk demos in the browser and visualize them. It already includes two demos of wireheading and noise-seeking agents going wrong.