RSS

Superrationality

TagLast edit: 17 Nov 2022 5:30 UTC by Roman Leventov

Superrationality is a concept invented by Douglas Hofstadter. He thought that agents should cooperate in Prisoner’s Dilemma, but the primary notion of “rationality” which had been deeply developed by economists, decision-theorists, and game-theorists disagreed. Rather than fighting over the definition of rational, Douglas Hofstadter coined the term superrational for the kind of rationality he was interested in.

Douglas Hofstadter did not publish a full theory of superrationality, instead focusing on the prisoner’s dilemma and other specific examples to communicate his intuition. However, some principles can be clearly inferred from his writing, such as reasoning as if you have control over the other player’s policy (in symmetric games).

Eliezer Yudkowsky shared the same core intuition with Douglas Hofstadter, but took the path of trying to reclaim the word rational for what he meant, in Functional Decision Theory. As a result, LessWrong does not consistently use superrational/​superrationality.

Su­per­ra­tional­ity and net­work flow control

Alex Flint22 Jul 2013 1:49 UTC
29 points
11 comments1 min readLW link

Su­per­ra­tional­ity in ar­bi­trary games

Vanessa Kosoy4 Nov 2015 18:20 UTC
31 points
6 comments6 min readLW link

Hofs­tadter’s Superrationality

gwern21 Apr 2012 13:33 UTC
70 points
21 comments1 min readLW link

LINK: Su­per­ra­tional­ity and DAOs

somnicule24 Jan 2015 9:47 UTC
8 points
10 comments1 min readLW link

Break­ing Or­a­cles: su­per­ra­tional­ity and acausal trade

Stuart_Armstrong25 Nov 2019 10:40 UTC
25 points
15 comments1 min readLW link

Su­per­ra­tional Agents Kelly Bet In­fluence!

abramdemski16 Apr 2021 22:08 UTC
47 points
7 comments5 min readLW link

Con­di­tions for Su­per­ra­tional­ity-mo­ti­vated Co­op­er­a­tion in a one-shot Pri­soner’s Dilemma

Jim Buhler19 Dec 2022 15:00 UTC
24 points
4 comments5 min readLW link

In­fra-Bayesian haggling

hannagabor20 May 2024 12:23 UTC
18 points
0 comments20 min readLW link

Co­op­er­a­tion is op­ti­mal, with weaker agents too  -  tldr

Ryo 18 Apr 2024 15:03 UTC
12 points
22 comments4 min readLW link
(medium.com)

Mul­ti­verse-wide Co­op­er­a­tion via Cor­re­lated De­ci­sion Making

Kaj_Sotala20 Aug 2017 12:01 UTC
5 points
2 comments1 min readLW link
(foundational-research.org)

Three rea­sons to cooperate

paulfchristiano24 Dec 2022 17:40 UTC
82 points
14 comments10 min readLW link
(sideways-view.com)

A Bench­mark for De­ci­sion Theories

StrivingForLegibility11 Jan 2024 18:54 UTC
10 points
0 comments2 min readLW link
No comments.