AI ALIGNMENT FORUMStuart_Armstrong
AF

User Profile

Ω661327

Recent Posts

Curated Posts
starCurated - Recent, high quality posts selected by the LessWrong moderation team.
help Learn More
Frontpage Posts
supervisor_accountPosts meeting our frontpage guidelines: aim to explain, not to persuade. Avoid meta-discussion
(includes curated content and frontpage posts)
help Learn More
All Posts
personIncludes personal and meta blogposts (as well as curated and frontpage).
help Learn More

Standard ML Oracles vs Counterfactual ones

Stuart_Armstrong5d5 points5 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
0

Bridging syntax and semantics with Quine's Gavagai

Stuart_Armstrong21d5 points1 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
1

Wireheading as a potential problem with the new impact measure

Stuart_Armstrong20d8 points4 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
19

Disagreement with Paul: alignment induction

Stuart_Armstrong1mo3 points1 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
4

Web of connotations: Bleggs, Rubes, thermostats and beliefs

Stuart_Armstrong1mo3 points7 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
0

Using expected utility for Good(hart)

Stuart_Armstrong2mo6 points6 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
1

Petrov corrigibility

Stuart_Armstrong1mo3 points1 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
9

Bridging syntax and semantics, empirically

Stuart_Armstrong1mo1 point5 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
0

Boltzmann brain decision theory

Stuart_Armstrong1mo3 points6 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
0

Corrigibility doesn't always have a good action to take

Stuart_Armstrong2mo4 points1 min readShow Highlightsubdirectory_arrow_left
brightness_1
Read Status
0
Load More...  

Recent Comments