Par­tially Step­ping Down Isolation

jefftkJul 8, 2020, 11:40 AM
18 points
10 comments2 min readLW link
(www.jefftk.com)

[AN #109]: Teach­ing neu­ral nets to gen­er­al­ize the way hu­mans would

Rohin ShahJul 22, 2020, 5:10 PM
17 points
3 comments9 min readLW link
(mailchi.mp)

Ag­gre­gat­ing forecasts

JsevillamolJul 23, 2020, 6:04 PM
17 points
4 comments4 min readLW link

[Question] Is there an easy way to turn a LW se­quence into an epub?

ChristianKlJul 18, 2020, 6:20 PM
17 points
9 comments1 min readLW link

June 2020 gw­ern.net newsletter

gwernJul 2, 2020, 2:19 PM
16 points
0 commentsLW link
(www.gwern.net)

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

Why is the im­pact penalty time-in­con­sis­tent?

Stuart_ArmstrongJul 9, 2020, 5:26 PM
16 points
1 comment2 min readLW link

[Question] Could Nixon go­ing to China be a cause for the big stag­na­tion?

ChristianKlJul 5, 2020, 6:58 AM
16 points
4 comments1 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)

Fea­tured Tags: July 2020

RubyJul 9, 2020, 5:00 PM
16 points
2 comments1 min readLW link

Causal­ity and its harms

George3d6Jul 4, 2020, 2:42 PM
16 points
19 comments11 min readLW link
(blog.cerebralab.com)

Roll for Sanity

eapacheJul 13, 2020, 4:39 PM
16 points
2 comments4 min readLW link

Weak HCH ac­cesses EXP

evhubJul 22, 2020, 10:36 PM
16 points
0 comments3 min readLW link

Sun­day Au­gust 2, 12pm (PDT) — talks by jim­ran­domh, john­swen­th­worth, Daniel Filan, Jacobian

Jul 30, 2020, 11:55 PM
15 points
2 comments1 min readLW link

UML IV: Lin­ear Predictors

Rafael HarthJul 8, 2020, 7:06 PM
15 points
0 comments9 min readLW link

Open & Wel­come Thread—July 2020

habrykaJul 2, 2020, 10:41 PM
15 points
80 comments1 min readLW link

Chains, Bot­tle­necks and Optimization

curiJul 21, 2020, 2:07 AM
14 points
12 comments4 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

On­line Meetup: Balaji Srini­vasan on “Na­tion of Emi­grants”

JoshuaFoxJul 22, 2020, 2:35 PM
14 points
1 comment1 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

Spoiler-Free Re­view: Witcher 3: Wild Hunt (plus a Spoiler­ific sec­tion)

ZviJul 4, 2020, 11:10 PM
14 points
2 comments17 min readLW link
(thezvi.wordpress.com)

Science eats its young

George3d6Jul 12, 2020, 12:32 PM
14 points
13 comments10 min readLW link
(cerebralab.com)

Ineffi­cient doesn’t mean in­differ­ent, but it might mean wimpy.

DirectedEvolutionJul 20, 2020, 6:27 PM
14 points
3 comments5 min readLW link

Diver­gence causes iso­lated de­mands for rigor

George3d6Jul 15, 2020, 6:59 PM
14 points
4 comments7 min readLW link
(blog.cerebralab.com)

[AN #106]: Eval­u­at­ing gen­er­al­iza­tion abil­ity of learned re­ward models

Rohin ShahJul 1, 2020, 5:20 PM
14 points
2 comments11 min readLW link
(mailchi.mp)

Goals and short descriptions

Michele CampoloJul 2, 2020, 5:41 PM
14 points
8 comments5 min readLW link

[Question] What do we now know about long-term con­se­quences of a COVID-19 in­fec­tion?

ChristianKlJul 15, 2020, 2:42 PM
14 points
1 comment1 min readLW link

The “best pre­dic­tor is mal­i­cious op­ti­miser” problem

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

[AN #107]: The con­ver­gent in­stru­men­tal sub­goals of goal-di­rected agents

Rohin ShahJul 16, 2020, 6:47 AM
13 points
1 comment8 min readLW link
(mailchi.mp)

[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)

[Question] Harry Pot­ter and meth­ods of ra­tio­nal­ity al­ter­na­tive end­ing.

Klen SalubriJul 1, 2020, 6:51 PM
13 points
25 comments1 min readLW link

Split­ting De­bate up into Two Subsystems

NandiJul 3, 2020, 8:11 PM
13 points
5 comments4 min readLW link

Meta-prefer­ences are weird

Jul 16, 2020, 11:03 PM
13 points
2 comments5 min readLW link

[Question] Repli­cated cog­ni­tive bias list?

moderockJul 4, 2020, 4:15 AM
12 points
3 comments1 min readLW link

No Ul­ti­mate Goal and a Small Ex­is­ten­tial Crisis

Logan RiggsJul 24, 2020, 6:39 PM
12 points
23 comments3 min readLW link

[Question] How big of an is­sue are patent trolls to the av­er­age startup?

ChristianKlJul 16, 2020, 11:31 AM
12 points
4 comments1 min readLW link

Re­search agenda for AI safety and a bet­ter civilization

agilecavemanJul 22, 2020, 6:35 AM
12 points
2 comments16 min readLW link

Restocking

jefftkJul 23, 2020, 1:40 PM
12 points
0 comments1 min readLW link
(www.jefftk.com)

Bot­tle­neck Examples

curiJul 23, 2020, 6:23 PM
11 points
0 comments9 min readLW link

The Echo Fallacy

Mati_RoyJul 5, 2020, 11:00 PM
11 points
3 comments1 min readLW link

Cal­ibrate words, not just probabilities

MikkWJul 18, 2020, 5:56 AM
11 points
3 comments2 min readLW link

Franken­stein De­lenda Est

eapacheJul 10, 2020, 2:38 PM
11 points
0 comments4 min readLW link

Pulse and Glide Cycling

jefftkJul 19, 2020, 7:02 PM
11 points
5 comments2 min readLW link
(www.jefftk.com)

As­set Prices Con­sis­tently Vio­late Effi­cient Mar­ket Hypothesis

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

[Question] [Re­source Re­quest] What are good re­sources for best prac­tice for cre­at­ing bu­reau­cracy?

ChristianKlJul 9, 2020, 12:05 PM
11 points
1 comment1 min readLW link

Meta Pro­gram­ming GPT: A route to Su­per­in­tel­li­gence?

dmteaJul 11, 2020, 2:51 PM
10 points
7 comments4 min readLW link

Spoiler-Free Re­view: Hori­zon Zero Dawn

ZviJul 7, 2020, 1:20 PM
10 points
1 comment7 min readLW link
(thezvi.wordpress.com)

[Question] Us­ing books to prime behavior

ryan wongJul 25, 2020, 5:59 AM
10 points
8 comments1 min readLW link

Par­allels Between AI Safety by De­bate and Ev­i­dence Law

CullenJul 20, 2020, 10:52 PM
10 points
1 comment2 min readLW link
(cullenokeefe.com)

[Question] (an­swered: yes) Has any­one writ­ten up a con­sid­er­a­tion of Downs’s “Para­dox of Vot­ing” from the per­spec­tive of MIRI-ish de­ci­sion the­o­ries (UDT, FDT, or even just EDT)?

Jameson QuinnJul 6, 2020, 6:26 PM
10 points
24 comments1 min readLW link