RSS

Myopia

TagLast edit: Dec 30, 2024, 11:19 AM by Dakara

Myopia refers to short-sightedness in planning and decision-making processes. It describes a tendency to prioritize immediate or short-term outcomes while disregarding longer-term consequences.

The most extreme form of myopia occurs when an agent considers only immediate rewards, completely disregarding future consequences. In artificial intelligence contexts, a perfectly myopic agent would optimize solely for the current query or task without attempting to influence future outcomes.

Myopic agents demonstrate several notable properties:

Par­tial Agency

abramdemskiSep 27, 2019, 10:04 PM
75 points
18 comments9 min readLW link

The Credit As­sign­ment Problem

abramdemskiNov 8, 2019, 2:50 AM
103 points
40 comments17 min readLW link1 review

How LLMs are and are not myopic

janusJul 25, 2023, 2:19 AM
134 points
16 comments8 min readLW link

Towards a mechanis­tic un­der­stand­ing of corrigibility

evhubAug 22, 2019, 11:20 PM
47 points
26 comments4 min readLW link

Open Prob­lems with Myopia

Mar 10, 2021, 6:38 PM
66 points
16 comments8 min readLW link

Steer­ing Be­havi­our: Test­ing for (Non-)My­opia in Lan­guage Models

Dec 5, 2022, 8:28 PM
40 points
19 comments10 min readLW link

LCDT, A My­opic De­ci­sion Theory

Aug 3, 2021, 10:41 PM
57 points
50 comments15 min readLW link

Defin­ing Myopia

abramdemskiOct 19, 2019, 9:32 PM
32 points
18 comments8 min readLW link

Ar­gu­ments against my­opic training

Richard_NgoJul 9, 2020, 4:07 PM
62 points
39 comments12 min readLW link

You can still fetch the coffee to­day if you’re dead tomorrow

davidadDec 9, 2022, 2:06 PM
96 points
19 comments5 min readLW link

2019 Re­view Rewrite: Seek­ing Power is Often Ro­bustly In­stru­men­tal in MDPs

TurnTroutDec 23, 2020, 5:16 PM
35 points
0 comments4 min readLW link
(www.lesswrong.com)

An overview of 11 pro­pos­als for build­ing safe ad­vanced AI

evhubMay 29, 2020, 8:38 PM
220 points
36 comments38 min readLW link2 reviews

Bayesian Evolv­ing-to-Extinction

abramdemskiFeb 14, 2020, 11:55 PM
40 points
13 comments5 min readLW link

Ran­dom Thoughts on Pre­dict-O-Matic

abramdemskiOct 17, 2019, 11:39 PM
35 points
3 comments9 min readLW link

The Parable of Pre­dict-O-Matic

abramdemskiOct 15, 2019, 12:49 AM
354 points
43 comments14 min readLW link2 reviews

Self-Fulfilling Prophe­cies Aren’t Always About Self-Awareness

John_MaxwellNov 18, 2019, 11:11 PM
14 points
7 comments4 min readLW link

The Dual­ist Pre­dict-O-Matic ($100 prize)

John_MaxwellOct 17, 2019, 6:45 AM
19 points
35 comments5 min readLW link

Why GPT wants to mesa-op­ti­mize & how we might change this

John_MaxwellSep 19, 2020, 1:48 PM
55 points
33 comments9 min readLW link

Seek­ing Power is Often Con­ver­gently In­stru­men­tal in MDPs

Dec 5, 2019, 2:33 AM
162 points
39 comments17 min readLW link2 reviews
(arxiv.org)

Un­der­stand­ing and con­trol­ling auto-in­duced dis­tri­bu­tional shift

L Rudolf LDec 13, 2021, 2:59 PM
33 points
4 comments16 min readLW link

Evan Hub­inger on Ho­mo­gene­ity in Take­off Speeds, Learned Op­ti­miza­tion and Interpretability

Michaël TrazziJun 8, 2021, 7:20 PM
28 points
0 comments55 min readLW link

Graph­i­cal World Models, Coun­ter­fac­tu­als, and Ma­chine Learn­ing Agents

Koen.HoltmanFeb 17, 2021, 11:07 AM
6 points
2 comments10 min readLW link

Non-my­opia stories

lberglundNov 13, 2023, 5:52 PM
29 points
10 comments7 min readLW link

Trans­form­ing my­opic op­ti­miza­tion to or­di­nary op­ti­miza­tion—Do we want to seek con­ver­gence for my­opic op­ti­miza­tion prob­lems?

tailcalledDec 11, 2021, 8:38 PM
12 points
1 comment5 min readLW link

GPT-4 al­ign­ing with aca­sual de­ci­sion the­ory when in­structed to play games, but in­cludes a CDT ex­pla­na­tion that’s in­cor­rect if they differ

Christopher KingMar 23, 2023, 4:16 PM
7 points
4 comments8 min readLW link

Un­der­speci­fi­ca­tion of Or­a­cle AI

Jan 15, 2023, 8:10 PM
30 points
12 comments19 min readLW link

How com­plex are my­opic imi­ta­tors?

Vivek HebbarFeb 8, 2022, 12:00 PM
26 points
1 comment15 min readLW link

AI safety via mar­ket making

evhubJun 26, 2020, 11:07 PM
71 points
45 comments1 min readLW link

In­ter­pretabil­ity’s Align­ment-Solv­ing Po­ten­tial: Anal­y­sis of 7 Scenarios

Evan R. MurphyMay 12, 2022, 8:01 PM
58 points
0 comments59 min readLW link

Ac­cept­abil­ity Ver­ifi­ca­tion: A Re­search Agenda

Jul 12, 2022, 8:11 PM
50 points
0 comments1 min readLW link
(docs.google.com)

Laz­i­ness in AI

Richard HenageSep 2, 2022, 5:04 PM
13 points
5 comments1 min readLW link

Gen­er­a­tive, Epi­sodic Ob­jec­tives for Safe AI

Michael GlassOct 5, 2022, 11:18 PM
11 points
3 comments8 min readLW link

Limit­ing an AGI’s Con­text Temporally

EulersApprenticeFeb 17, 2019, 3:29 AM
5 points
11 comments1 min readLW link

Simulators

janusSep 2, 2022, 12:45 PM
631 points
168 comments41 min readLW link8 reviews
(generative.ink)

Fight­ing Akra­sia: In­cen­tivis­ing Action

Gordon Seidoh WorleyApr 29, 2009, 1:48 PM
12 points
58 comments2 min readLW link

GPT-4 busted? Clear self-in­ter­est when sum­ma­riz­ing ar­ti­cles about it­self vs when ar­ti­cle talks about Claude, LLaMA, or DALL·E 2

Christopher KingMar 31, 2023, 5:05 PM
6 points
4 comments4 min readLW link
No comments.