Build­ing up to an In­ter­nal Fam­ily Sys­tems model

Kaj_SotalaJan 26, 2019, 12:25 PM
289 points
86 comments28 min readLW link2 reviews

Fu­ture di­rec­tions for nar­row value learning

Rohin ShahJan 26, 2019, 2:36 AM
12 points
4 comments4 min readLW link

[Question] For what do we need Su­per­in­tel­li­gent AI?

avturchinJan 25, 2019, 3:01 PM
14 points
18 comments1 min readLW link

“AlphaS­tar: Mas­ter­ing the Real-Time Strat­egy Game StarCraft II”, Deep­Mind [won 10 of 11 games against hu­man pros]

gwernJan 24, 2019, 8:49 PM
62 points
52 comments1 min readLW link
(deepmind.com)

Thoughts on re­ward en­g­ineer­ing

paulfchristianoJan 24, 2019, 8:15 PM
30 points
30 comments11 min readLW link

From Per­sonal to Pri­son Gangs: En­forc­ing Proso­cial Behavior

johnswentworthJan 24, 2019, 6:07 PM
153 points
26 comments5 min readLW link2 reviews

[Question] Is Agent Si­mu­lates Pre­dic­tor a “fair” prob­lem?

Chris_LeongJan 24, 2019, 1:18 PM
22 points
19 comments1 min readLW link

[Question] In what way has the gen­er­a­tion af­ter us “gone too far”?

EloJan 24, 2019, 10:22 AM
8 points
3 comments1 min readLW link

The hu­man side of interaction

Rohin ShahJan 24, 2019, 10:14 AM
19 points
5 comments4 min readLW link

South Bay Meetup

David FriedmanJan 24, 2019, 5:29 AM
1 point
0 commentsLW link

Allow­ing a for­mal proof sys­tem to self im­prove while avoid­ing Lo­bian ob­sta­cles.

Donald HobsonJan 23, 2019, 11:04 PM
6 points
4 comments2 min readLW link

Ra­tion­al­ity Café No. 9 - The Se­quences, Part 1; Sec­tion C b

thegreatnickJan 23, 2019, 10:41 PM
1 point
0 comments1 min readLW link

Ed­in­burgh SSC meetup ϛ

Hamish Peter ToddJan 23, 2019, 6:36 PM
1 point
0 commentsLW link

S-Curves for Trend Forecasting

Matt GoldenbergJan 23, 2019, 6:17 PM
113 points
23 comments7 min readLW link4 reviews

One-step hy­po­thet­i­cal preferences

Stuart_ArmstrongJan 23, 2019, 3:14 PM
10 points
6 comments4 min readLW link

Shang­hai LessWrong Meetup

MindeyJan 23, 2019, 12:59 PM
11 points
10 comments1 min readLW link

Games in Kocherga club: Fal­la­cy­ma­nia, Tower of Chaos, Scien­tific Discovery

Alexander230Jan 23, 2019, 11:57 AM
10 points
0 comments1 min readLW link

Could de­clin­ing in­ter­est to the Dooms­day Ar­gu­ment ex­plain the Dooms­day Ar­gu­ment?

avturchinJan 23, 2019, 11:51 AM
8 points
3 comments2 min readLW link

Link: That Time a Guy Tried to Build a Utopia for Mice and it all Went to Hell

AlexeiJan 23, 2019, 6:27 AM
15 points
6 comments1 min readLW link

Learn­ing with catastrophes

paulfchristianoJan 23, 2019, 3:01 AM
27 points
9 comments4 min readLW link

The Re­la­tion­ship Between Hier­ar­chy and Wealth

sarahconstantinJan 23, 2019, 2:00 AM
59 points
8 comments12 min readLW link
(srconstantin.wordpress.com)

Too Smart for My Own Good

isovectorJan 22, 2019, 5:51 PM
38 points
4 comments3 min readLW link

Vote count­ing bug?

Richard_NgoJan 22, 2019, 3:44 PM
6 points
8 comments1 min readLW link

Stale air /​ high CO2 may de­crease your cog­ni­tive function

Yoav RavidJan 22, 2019, 12:52 PM
16 points
6 comments1 min readLW link

Why don’t peo­ple use for­mal meth­ods?

John_MaxwellJan 22, 2019, 9:39 AM
20 points
5 comments1 min readLW link
(www.hillelwayne.com)

[Question] Should ques­tions be called “ques­tions” or “con­fu­sions” (or “other”)?

RaemonJan 22, 2019, 2:45 AM
16 points
4 comments1 min readLW link

Align­ment Newslet­ter #42

Rohin ShahJan 22, 2019, 2:00 AM
20 points
1 comment10 min readLW link
(mailchi.mp)

Pitts­burgh SSC Meetup

fr8trainJan 22, 2019, 12:04 AM
1 point
0 comments1 min readLW link

Jan­uary 2019 Nashville SSC Meetup

Dude McDudeJan 21, 2019, 9:51 PM
3 points
3 comments1 min readLW link

Game Anal­y­sis Index

ZviJan 21, 2019, 3:30 PM
12 points
0 comments1 min readLW link
(thezvi.wordpress.com)

Disen­tan­gling ar­gu­ments for the im­por­tance of AI safety

Richard_NgoJan 21, 2019, 12:41 PM
133 points
23 comments8 min readLW link

Do the best ideas float to the top?

QuinnJan 21, 2019, 5:22 AM
22 points
11 comments6 min readLW link

Fol­low­ing hu­man norms

Rohin ShahJan 20, 2019, 11:59 PM
30 points
10 comments5 min readLW link

An­nounce­ment: AI al­ign­ment prize round 4 winners

cousin_itJan 20, 2019, 2:46 PM
74 points
41 comments1 min readLW link

Ca­pa­bil­ity amplification

paulfchristianoJan 20, 2019, 7:03 AM
24 points
8 comments13 min readLW link

Less Com­pe­ti­tion, More Mer­i­toc­racy?

ZviJan 20, 2019, 2:00 AM
85 points
19 comments20 min readLW link3 reviews
(thezvi.wordpress.com)

[Question] Why not tool AI?

smitheeJan 19, 2019, 10:18 PM
20 points
11 comments1 min readLW link

[Question] What math do i need for data anal­y­sis?

Yoav RavidJan 19, 2019, 10:08 AM
13 points
13 comments1 min readLW link

Re­ward uncertainty

Rohin ShahJan 19, 2019, 2:16 AM
26 points
3 comments5 min readLW link

Do­ing De­spite Dis­lik­ing: Self‐reg­u­la­tory Strate­gies in Every­day Aver­sive Activities

lifelonglearnerJan 19, 2019, 12:27 AM
14 points
1 comment1 min readLW link
(onlinelibrary.wiley.com)

Bun­dle your Experiments

PeterMcCluskeyJan 18, 2019, 11:22 PM
19 points
0 comments2 min readLW link
(www.bayesianinvestor.com)

Cur­ing the World of Men

Jacob FalkovichJan 18, 2019, 8:23 PM
−7 points
4 comments1 min readLW link
(putanumonit.com)

How could shares in a megapro­ject re­turn value to share­hold­ers?

ryan_bJan 18, 2019, 6:36 PM
17 points
12 comments1 min readLW link

Syn­the­sis­ing di­ver­gent prefer­ences: an ex­am­ple in pop­u­la­tion ethics

Stuart_ArmstrongJan 18, 2019, 2:29 PM
12 points
8 comments4 min readLW link

The Very Repug­nant Conclusion

Stuart_ArmstrongJan 18, 2019, 2:26 PM
26 points
19 comments1 min readLW link

EA Funds: Long-Term Fu­ture fund is open to ap­pli­ca­tions un­til Feb. 7th

habrykaJan 17, 2019, 8:27 PM
24 points
2 comments1 min readLW link

Fi­nance Followups

Jacob FalkovichJan 17, 2019, 5:54 PM
26 points
8 comments4 min readLW link

De­bate AI and the De­ci­sion to Re­lease an AI

Chris_LeongJan 17, 2019, 2:36 PM
9 points
18 comments3 min readLW link

An­throp­ics is pretty normal

Stuart_ArmstrongJan 17, 2019, 1:26 PM
39 points
9 comments8 min readLW link

Solv­ing the Dooms­day argument

Stuart_ArmstrongJan 17, 2019, 12:32 PM
14 points
16 comments1 min readLW link