RSS

Roland Pihlakas

Karma: 67

Why mod­el­ling multi-ob­jec­tive home­osta­sis is es­sen­tial for AI al­ign­ment (and how it helps with AI safety as well)

Roland Pihlakas12 Jan 2025 3:37 UTC
38 points
5 comments10 min readLW link

Build­ing AI safety bench­mark en­vi­ron­ments on themes of uni­ver­sal hu­man values

Roland Pihlakas3 Jan 2025 4:24 UTC
17 points
3 comments8 min readLW link
(docs.google.com)

Sets of ob­jec­tives for a multi-ob­jec­tive RL agent to optimize

Ben Smith and Roland Pihlakas
23 Nov 2022 6:49 UTC
13 points
0 comments8 min readLW link

A brief re­view of the rea­sons multi-ob­jec­tive RL could be im­por­tant in AI Safety Research

Ben Smith and Roland Pihlakas
29 Sep 2021 17:09 UTC
30 points
7 comments10 min readLW link