Coun­ter­fac­tu­als are an An­swer, Not a Question

Chris_LeongSep 3, 2019, 3:36 PM
14 points
6 comments4 min readLW link

How to write good AI fore­cast­ing ques­tions + Ques­tion Database (Fore­cast­ing in­fras­truc­ture, part 3)

Sep 3, 2019, 2:50 PM
29 points
0 comments4 min readLW link

How Speci­fic­ity Works

LironSep 3, 2019, 12:11 PM
90 points
47 comments7 min readLW link

[Ham­mer­time Fi­nal Exam] Quan­tum Walk, Or­a­cles and Sunk Meaning

silentbobSep 3, 2019, 11:59 AM
22 points
4 comments5 min readLW link

AIXSU—AI and X-risk Strat­egy Unconference

David_KristofferssonSep 3, 2019, 11:35 AM
24 points
3 comments2 min readLW link

Book Re­view: Ages Of Discord

Scott AlexanderSep 3, 2019, 6:30 AM
36 points
1 comment21 min readLW link
(slatestarcodex.com)

Open & Wel­come Thread—Septem­ber 2019

RubySep 3, 2019, 2:53 AM
9 points
37 comments1 min readLW link

The Trans­par­ent So­ciety: A rad­i­cal trans­for­ma­tion that we should prob­a­bly undergo

mako yassSep 3, 2019, 2:27 AM
12 points
26 comments8 min readLW link

One Way to Think About ML Transparency

Matthew BarnettSep 2, 2019, 11:27 PM
26 points
28 comments5 min readLW link

Joy in Dis­cov­ery: Galois theory

Eigil RischelSep 2, 2019, 7:16 PM
30 points
2 comments3 min readLW link

Sto­ries About Academia

Ben PaceSep 2, 2019, 6:40 PM
34 points
1 comment19 min readLW link

Best util­ity nor­mal­i­sa­tion method to date?

Stuart_ArmstrongSep 2, 2019, 6:24 PM
19 points
7 comments3 min readLW link

South Bay Meetup, San Jose, CA

DavidFriedmanSep 2, 2019, 6:23 PM
2 points
0 commentsLW link

[Question] What hap­pened to Lev­er­age Re­search?

ChristianKlSep 2, 2019, 5:45 PM
19 points
6 comments1 min readLW link

The Power to De­mol­ish Bad Arguments

LironSep 2, 2019, 12:57 PM
100 points
84 comments11 min readLW link6 reviews

Speci­fic­ity: Your Brain’s Superpower

LironSep 2, 2019, 12:53 PM
85 points
8 comments1 min readLW link

How Much is Your Time Worth?

lynettebyeSep 2, 2019, 6:19 AM
79 points
22 comments6 min readLW link1 review

Kate Stil­lwell Talk—Berkeley Effec­tive Altru­ism Meetup at REACH

hamnoxSep 1, 2019, 8:12 PM
0 points
0 comments1 min readLW link

Ra­tion­al­ity Meetup at REACH (Grat­i­tudes)

hamnoxSep 1, 2019, 8:07 PM
1 point
0 comments1 min readLW link

REACH Read­ing Group: Soviet Bioweapons Program

hamnoxSep 1, 2019, 7:54 PM
2 points
0 comments1 min readLW link

[Question] What are the biggest “moon­shots” cur­rently in progress?

LironSep 1, 2019, 7:41 PM
14 points
14 comments1 min readLW link

Effec­tive Altru­ist Group De­bug­ging at REACH

hamnoxSep 1, 2019, 7:33 PM
1 point
0 comments1 min readLW link

Prob­a­bil­ity as Min­i­mal Map

johnswentworthSep 1, 2019, 7:19 PM
52 points
10 comments6 min readLW link

Tiddly­wiki for or­ga­niz­ing notes and research

eukaryoteSep 1, 2019, 6:44 PM
42 points
8 comments4 min readLW link
(eukaryotewritesblog.com)

Au­gust 2019 gw­ern.net newslet­ter (pop­ups.js demo)

gwernSep 1, 2019, 5:52 PM
12 points
6 commentsLW link
(www.gwern.net)

[Question] Ques­tion about a past donor to MIRI.

Teerth AlokeSep 1, 2019, 3:39 PM
2 points
4 comments1 min readLW link

The Five Main Mus­cles for a Full Range of Nat­u­ral Move­ment, Dy­namic Align­ment & Balance.

leggiSep 1, 2019, 3:22 AM
15 points
10 comments10 min readLW link

Ar­gu­ing Ab­solute Velocities

Closed Limelike CurvesSep 1, 2019, 2:11 AM
13 points
2 comments4 min readLW link

Bay Area Win­ter Sols­tice Volunteering

Chelsea VossSep 1, 2019, 2:05 AM
28 points
0 comments1 min readLW link

Log­i­cal Coun­ter­fac­tu­als and Propo­si­tion graphs, Part 2

Donald HobsonAug 31, 2019, 8:58 PM
13 points
0 comments3 min readLW link

Cam­bridge LW/​SSC Meetup

NoSignalNoNoiseAug 31, 2019, 5:10 PM
6 points
0 comments1 min readLW link

Peter Thiel/​Eric We­in­stein Tran­script on Growth, Violence, and Stories

Ben PaceAug 31, 2019, 2:44 AM
70 points
20 comments15 min readLW link

[Question] What does so­cial psy­chol­ogy tell us about get­ting con­sis­tent eval­u­a­tions out of groups?

ElizabethAug 31, 2019, 1:11 AM
22 points
0 comments1 min readLW link

[Question] How good a proxy for ac­cu­racy is pre­ci­sion?

ElizabethAug 31, 2019, 1:01 AM
16 points
5 comments1 min readLW link

[Question] Can this model grade a test with­out know­ing the an­swers?

ElizabethAug 31, 2019, 12:53 AM
20 points
3 comments1 min readLW link

Look at the Shape of Your Utility Distribution

zulupineappleAug 30, 2019, 11:27 PM
15 points
1 comment3 min readLW link

[Question] Does any­one know of a good overview of what hu­mans know about de­pres­sion?

Eli TyreAug 30, 2019, 11:22 PM
13 points
2 comments1 min readLW link

Septem­ber Brag­ging Thread

RaemonAug 30, 2019, 9:58 PM
47 points
12 comments1 min readLW link

2-D Robustness

Vlad MikulikAug 30, 2019, 8:27 PM
85 points
8 comments2 min readLW link

How to Make Billions of Dol­lars Re­duc­ing Loneliness

John_MaxwellAug 30, 2019, 5:30 PM
60 points
32 comments7 min readLW link

A Game of Gi­ants [Wait But Why]

kilotarasAug 30, 2019, 5:48 AM
5 points
4 comments1 min readLW link
(waitbutwhy.com)

The Cri­sis of Ev­i­dence Use

Ian David MossAug 30, 2019, 4:50 AM
2 points
0 comments4 min readLW link
(medium.com)

AI Align­ment Writ­ing Day Roundup #1

Ben PaceAug 30, 2019, 1:26 AM
32 points
12 comments1 min readLW link

Min­neapo­lis SSC meetup Septem­ber 2019

Drake ThomasAug 29, 2019, 11:01 PM
1 point
0 comments1 min readLW link

LW Team Up­dates—Septem­ber 2019

RubyAug 29, 2019, 10:12 PM
39 points
13 comments2 min readLW link

Why study per­tur­ba­tive ad­ver­sar­ial at­tacks ?

Yuxi_LiuAug 29, 2019, 8:15 PM
19 points
1 comment2 min readLW link

AI Fore­cast­ing Re­s­olu­tion Coun­cil (Fore­cast­ing in­fras­truc­ture, part 2)

Aug 29, 2019, 5:35 PM
35 points
2 comments3 min readLW link

[Question] Why are the peo­ple who could be do­ing safety re­search, but aren’t, do­ing some­thing else?

Adam SchollAug 29, 2019, 8:51 AM
27 points
19 comments1 min readLW link

[Link] Book Re­view: Refram­ing Su­per­in­tel­li­gence (SSC)

ioannesAug 28, 2019, 10:57 PM
46 points
9 comments2 min readLW link

[Question] I think I came up with a good util­ity func­tion for AI that seems too ob­vi­ous. Can you peo­ple poke holes in it?

litten8Aug 28, 2019, 10:33 PM
1 point
7 comments1 min readLW link