Learn­ing the prior and generalization

evhubJul 29, 2020, 10:49 PM
34 points
16 comments4 min readLW link

What Failure Looks Like: Distill­ing the Discussion

Ben PaceJul 29, 2020, 9:49 PM
82 points
14 comments7 min readLW link

New Paper on Herd Im­mu­nity Thresholds

ZviJul 29, 2020, 8:50 PM
41 points
16 comments9 min readLW link
(thezvi.wordpress.com)

What a 20-year-lead in mil­i­tary tech might look like

Daniel KokotajloJul 29, 2020, 8:10 PM
76 points
55 comments16 min readLW link

En­gag­ing Se­ri­ously with Short Timelines

sapphireJul 29, 2020, 7:21 PM
43 points
21 comments3 min readLW link

[AN #110]: Learn­ing fea­tures from hu­man feed­back to en­able re­ward learning

Rohin ShahJul 29, 2020, 5:20 PM
13 points
2 comments10 min readLW link
(mailchi.mp)

The “best pre­dic­tor is mal­i­cious op­ti­miser” problem

Donald HobsonJul 29, 2020, 11:49 AM
14 points
10 comments2 min readLW link

Im­prov­ing lo­cal gov­er­nance in frag­ile states—prac­ti­cal les­sons from the field

Tim LiptrotJul 29, 2020, 1:54 AM
16 points
3 comments6 min readLW link

Pa­ram­e­ters of Privacy

RaemonJul 29, 2020, 1:18 AM
31 points
2 comments6 min readLW link

Pre­dic­tions for GPT-N

hippkeJul 29, 2020, 1:16 AM
36 points
31 comments1 min readLW link

Tag­ging Open Call /​ Dis­cus­sion Thread

RubyJul 28, 2020, 9:58 PM
65 points
118 comments4 min readLW link

Wiki-Tag FAQ

RubyJul 28, 2020, 9:57 PM
39 points
4 comments13 min readLW link

Tags Dis­cus­sion/​Talk Thread

RubyJul 28, 2020, 9:57 PM
30 points
81 comments1 min readLW link

[Question] What hap­pens to var­i­ance as neu­ral net­work train­ing is scaled? What does it im­ply about “lot­tery tick­ets”?

abramdemskiJul 28, 2020, 8:22 PM
25 points
4 comments1 min readLW link

[Question] How will in­ter­net fo­rums like LW be able to defend against GPT-style spam?

ChristianKlJul 28, 2020, 8:12 PM
14 points
17 comments1 min readLW link

[Question] To what ex­tent are the scal­ing prop­er­ties of Trans­former net­works ex­cep­tional?

abramdemskiJul 28, 2020, 8:06 PM
30 points
1 comment1 min readLW link

The Con­ceited Folly of Certainty

Noah BlaffJul 28, 2020, 7:56 PM
7 points
3 comments20 min readLW link

The Curse of Cur­sory Idealism

Noah BlaffJul 28, 2020, 7:56 PM
9 points
1 comment3 min readLW link

[Question] Does the lot­tery ticket hy­poth­e­sis sug­gest the scal­ing hy­poth­e­sis?

Daniel KokotajloJul 28, 2020, 7:52 PM
14 points
17 comments1 min readLW link

[Question] Prob­a­bil­ity that other ar­chi­tec­tures will scale as well as Trans­form­ers?

Daniel KokotajloJul 28, 2020, 7:36 PM
22 points
4 comments1 min readLW link

[Question] How should I back up and redo, in a pub­li­cly-ed­ited ar­ti­cle?

Jameson QuinnJul 28, 2020, 7:07 PM
7 points
1 comment1 min readLW link

Reread­ing At­las Shrugged

VaniverJul 28, 2020, 6:54 PM
174 points
36 comments13 min readLW link1 review

FHI Re­port: How Will Na­tional Se­cu­rity Con­sid­er­a­tions Affect An­titrust De­ci­sions in AI? An Ex­am­i­na­tion of His­tor­i­cal Precedents

CullenJul 28, 2020, 6:34 PM
2 points
0 comments1 min readLW link
(www.fhi.ox.ac.uk)

Del­e­gate a Forecast

AmandangoJul 28, 2020, 5:43 PM
44 points
25 comments2 min readLW link
(forum.effectivealtruism.org)

As­set Prices Con­sis­tently Vio­late Effi­cient Mar­ket Hypothesis

LironJul 28, 2020, 2:21 PM
11 points
3 comments1 min readLW link
(lt3000.blogspot.com)

A com­mu­nity-cu­rated repos­i­tory of in­ter­est­ing GPT-3 stuff

Rudi CJul 28, 2020, 2:16 PM
8 points
0 commentsLW link
(github.com)

[Question] Billion­aire Economics

Virgil KurkjianJul 28, 2020, 3:01 AM
22 points
31 comments1 min readLW link

[Question] What spe­cific dan­gers arise when ask­ing GPT-N to write an Align­ment Fo­rum post?

Matthew BarnettJul 28, 2020, 2:56 AM
45 points
14 comments1 min readLW link

The Fu­ture of Science

Richard_NgoJul 28, 2020, 2:43 AM
21 points
2 comments7 min readLW link

Ba­sic Con­ver­sa­tional Co­or­di­na­tion: Micro-co­or­di­na­tion of In­ten­tion

Eli TyreJul 27, 2020, 10:41 PM
41 points
7 comments2 min readLW link

[up­dated] how does gpt2′s train­ing cor­pus cap­ture in­ter­net dis­cus­sion? not well

nostalgebraistJul 27, 2020, 10:30 PM
25 points
3 comments2 min readLW link
(nostalgebraist.tumblr.com)

AI and Efficiency

DragonGodJul 27, 2020, 8:58 PM
9 points
1 comment1 min readLW link
(openai.com)

The Rise of Com­mon­sense Reasoning

DragonGodJul 27, 2020, 7:01 PM
8 points
0 commentsLW link
(www.reddit.com)

Why You Might Want a New Way to Vi­su­al­ize Biases

dtmJul 27, 2020, 5:30 PM
18 points
2 comments3 min readLW link

Are we in an AI over­hang?

Andy JonesJul 27, 2020, 12:48 PM
266 points
106 comments4 min readLW link

Gen­er­al­iz­ing the Power-Seek­ing Theorems

TurnTroutJul 27, 2020, 12:28 AM
41 points
6 comments4 min readLW link

Change the world a lit­tle bit

DirectedEvolutionJul 26, 2020, 10:35 PM
23 points
3 comments2 min readLW link

Devel­op­men­tal Stages of GPTs

orthonormalJul 26, 2020, 10:03 PM
140 points
72 comments7 min readLW link1 review

The prob­lem of Pop Analogy

thesilvermenJul 26, 2020, 9:44 PM
4 points
0 comments1 min readLW link

[Question] Where do peo­ple dis­cuss do­ing things with GPT-3?

skybrianJul 26, 2020, 2:31 PM
2 points
7 comments1 min readLW link

Jam is obsolete

jefftkJul 26, 2020, 2:00 PM
25 points
24 comments1 min readLW link
(www.jefftk.com)

How I use Google Forms in my Workflow

Harri BesceliJul 26, 2020, 12:17 PM
34 points
4 comments3 min readLW link

[Question] Godel in sec­ond-or­der logic?

Abhimanyu Pallavi SudhirJul 26, 2020, 7:16 AM
6 points
4 comments1 min readLW link

A Nat­u­ral Ex­pla­na­tion of Nash Equilibria

Dr. BirdbrainJul 25, 2020, 11:40 PM
3 points
6 comments5 min readLW link

A Scal­able Ur­ban De­sign and the Sin­gle Build­ing City

Edward SwernofskyJul 25, 2020, 7:06 PM
22 points
21 comments17 min readLW link

Con­straints from nat­u­ral­ized ethics.

Charlie SteinerJul 25, 2020, 2:54 PM
21 points
0 comments3 min readLW link

What it means to optimise

Neel NandaJul 25, 2020, 9:40 AM
6 points
0 comments8 min readLW link
(www.neelnanda.io)

Ac­cess to AI: a hu­man right?

dmteaJul 25, 2020, 9:38 AM
5 points
3 comments2 min readLW link

How to learn from conversations

Neel NandaJul 25, 2020, 9:36 AM
16 points
3 comments4 min readLW link
(www.neelnanda.io)

Tak­ing the first step

Neel NandaJul 25, 2020, 9:33 AM
10 points
0 comments16 min readLW link
(www.neelnanda.io)