Is Ra­tion­al­ist Self-Im­prove­ment Real?

Jacob FalkovichDec 9, 2019, 5:11 PM
263 points
78 comments11 min readLW link3 reviews

Moloch Hasn’t Won

ZviDec 28, 2019, 4:30 PM
188 points
40 comments7 min readLW link1 review
(thezvi.wordpress.com)

Paper-Read­ing for Gears

johnswentworthDec 4, 2019, 9:02 PM
165 points
6 comments4 min readLW link1 review

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 “Deep Dou­ble Des­cent”

evhubDec 6, 2019, 12:00 AM
151 points
51 comments5 min readLW link4 reviews

Prop­a­gat­ing Facts into Aesthetics

RaemonDec 19, 2019, 4:09 AM
134 points
38 comments11 min readLW link1 review

2019 AI Align­ment Liter­a­ture Re­view and Char­ity Comparison

LarksDec 19, 2019, 3:00 AM
130 points
18 comments62 min readLW link

Firm­ing Up Not-Ly­ing Around Its Edge-Cases Is Less Broadly Use­ful Than One Might Ini­tially Think

Zack_M_DavisDec 27, 2019, 5:09 AM
128 points
43 comments8 min readLW link2 reviews

We run the Cen­ter for Ap­plied Ra­tion­al­ity, AMA

AnnaSalamonDec 19, 2019, 4:34 PM
108 points
324 comments1 min readLW link

Karate Kid and Real­is­tic Ex­pec­ta­tions for Disagree­ment Resolution

RaemonDec 4, 2019, 11:25 PM
103 points
23 comments4 min readLW link

Against Pre­ma­ture Ab­strac­tion of Poli­ti­cal Issues

Wei DaiDec 18, 2019, 8:19 PM
94 points
22 comments2 min readLW link

Med­i­ta­tion Re­treat: Im­moral Mazes Se­quence Introduction

ZviDec 28, 2019, 12:50 AM
89 points
16 comments4 min readLW link
(thezvi.wordpress.com)

Bayesian examination

Lê Nguyên HoangDec 9, 2019, 7:50 PM
86 points
56 comments5 min readLW link

Re­cent Progress in the The­ory of Neu­ral Networks

intersticeDec 4, 2019, 11:11 PM
83 points
9 comments9 min readLW link

Ap­proval Ex­trac­tion Ad­ver­tised as Production

BenquoDec 15, 2019, 2:51 PM
82 points
39 comments19 min readLW link3 reviews
(benjaminrosshoffman.com)

Hu­mans Are Embed­ded Agents Too

johnswentworthDec 23, 2019, 7:21 PM
82 points
21 comments5 min readLW link

The Ac­tion­able Ver­sion of “Keep Your Iden­tity Small”

HazardDec 6, 2019, 1:34 AM
81 points
14 comments4 min readLW link

Free Speech and Triskaideka­pho­bic Calcu­la­tors: A Re­ply to Hub­inger on the Rele­vance of Public On­line Dis­cus­sion to Ex­is­ten­tial Risk

Zack_M_DavisDec 21, 2019, 12:49 AM
77 points
25 comments2 min readLW link

Pro­gram­mers Should Plan For Lower Pay

jefftkDec 29, 2019, 3:20 AM
77 points
72 comments2 min readLW link
(www.jefftk.com)

[Question] Un­der what cir­cum­stances is “don’t look at ex­ist­ing re­search” good ad­vice?

Kaj_SotalaDec 13, 2019, 1:59 PM
76 points
18 comments2 min readLW link

Im­perfect Competition

ZviDec 30, 2019, 12:20 PM
73 points
8 comments10 min readLW link
(thezvi.wordpress.com)

[Question] What de­ter­mines the bal­ance be­tween in­tel­li­gence sig­nal­ing and virtue sig­nal­ing?

Wei DaiDec 9, 2019, 12:11 AM
70 points
37 comments2 min readLW link

In­duc­tive bi­ases stick around

evhubDec 18, 2019, 7:52 PM
64 points
15 comments3 min readLW link

CFAR: Progress Re­port & Fu­ture Plans

Adam SchollDec 19, 2019, 6:19 AM
63 points
5 comments7 min readLW link

Confabulation

lsusrDec 8, 2019, 10:18 AM
62 points
6 comments2 min readLW link

[Part 2] Am­plify­ing gen­er­al­ist re­search via fore­cast­ing – re­sults from a pre­limi­nary exploration

Dec 19, 2019, 3:49 PM
62 points
10 comments14 min readLW link1 review

Pre­face to CLR’s Re­search Agenda on Co­op­er­a­tion, Con­flict, and TAI

JesseCliftonDec 13, 2019, 9:02 PM
62 points
10 comments2 min readLW link

One Million Dollars

jefftkDec 19, 2019, 2:10 AM
61 points
1 comment1 min readLW link
(www.jefftk.com)

Might hu­mans not be the most in­tel­li­gent an­i­mals?

Matthew BarnettDec 23, 2019, 9:50 PM
56 points
41 comments3 min readLW link

Pre­dic­tive cod­ing = RL + SL + Bayes + MPC

Steven ByrnesDec 10, 2019, 11:45 AM
56 points
8 comments8 min readLW link

MIRI’s 2019 Fundraiser

MaloDec 3, 2019, 1:16 AM
55 points
0 comments9 min readLW link

[Part 1] Am­plify­ing gen­er­al­ist re­search via fore­cast­ing – Models of im­pact and challenges

Dec 19, 2019, 3:50 PM
55 points
29 comments17 min readLW link

“You can’t pos­si­bly suc­ceed with­out [My Pet Is­sue]”

RaemonDec 19, 2019, 1:12 AM
54 points
14 comments1 min readLW link

Perfect Competition

ZviDec 29, 2019, 1:30 PM
54 points
29 comments7 min readLW link
(thezvi.wordpress.com)

The Re­view Phase

Ben PaceDec 9, 2019, 12:54 AM
53 points
3 comments5 min readLW link

LW For Ex­ter­nal Com­ments?

jefftkDec 5, 2019, 6:30 PM
52 points
7 comments1 min readLW link
(www.jefftk.com)

Speak­ing Truth to Power Is a Schel­ling Point

Zack_M_DavisDec 30, 2019, 6:12 AM
52 points
19 comments2 min readLW link

Gen­er­al­iz­ing Ex­per­i­men­tal Re­sults by Lev­er­ag­ing Knowl­edge of Mechanisms

Carlos_CinelliDec 11, 2019, 8:39 PM
50 points
5 comments1 min readLW link

Is the rate of sci­en­tific progress slow­ing down? (by Tyler Cowen and Ben South­wood)

Ben PaceDec 2, 2019, 3:45 AM
49 points
6 comments1 min readLW link
(docs.google.com)

T-Shaped Organizations

VaniverDec 16, 2019, 11:48 PM
49 points
6 comments2 min readLW link

2020′s Pre­dic­tion Thread

Ben GoldhaberDec 30, 2019, 11:18 PM
48 points
77 comments1 min readLW link

Self and No-Self

VaniverDec 29, 2019, 6:15 AM
48 points
3 comments2 min readLW link

Another AI Win­ter?

PeterMcCluskeyDec 25, 2019, 12:58 AM
47 points
14 comments4 min readLW link
(www.bayesianinvestor.com)

Should We Still Fly?

jefftkDec 20, 2019, 2:00 AM
47 points
13 comments1 min readLW link
(www.jefftk.com)

2010s Pre­dic­tions Review

ToflyDec 30, 2019, 9:47 PM
47 points
10 comments1 min readLW link

[Per­sonal Ex­per­i­ment] One Year with­out Junk Media

lsusrDec 14, 2019, 8:26 AM
45 points
11 comments1 min readLW link

A parable in the style of In­visi­ble Cities

Daniel KokotajloDec 16, 2019, 3:55 PM
45 points
6 comments1 min readLW link

Un­der­stand­ing Ma­chine Learn­ing (I)

Rafael HarthDec 20, 2019, 6:22 PM
44 points
12 comments11 min readLW link

A list of good heuris­tics that the case for AI x-risk fails

David Scott Krueger (formerly: capybaralet)Dec 2, 2019, 7:26 PM
43 points
15 comments2 min readLW link

A dilemma for pro­saic AI alignment

Daniel KokotajloDec 17, 2019, 10:11 PM
42 points
31 comments3 min readLW link