Re­laxed ad­ver­sar­ial train­ing for in­ner alignment

evhub10 Sep 2019 23:03 UTC
69 points
27 comments27 min readLW link

“AI-GAs: AI-gen­er­at­ing al­gorithms, an al­ter­nate paradigm for pro­duc­ing gen­eral ar­tifi­cial in­tel­li­gence”, Clune 2019

gwern10 Sep 2019 21:33 UTC
13 points
5 comments1 min readLW link
(arxiv.org)

SSC Mee­tups Every­where: Toulouse

Gyrodiot10 Sep 2019 19:17 UTC
3 points
0 comments1 min readLW link

[AN #63] How ar­chi­tec­ture search, meta learn­ing, and en­vi­ron­ment de­sign could lead to gen­eral intelligence

Rohin Shah10 Sep 2019 19:10 UTC
21 points
12 comments8 min readLW link
(mailchi.mp)

Free-to-Play Games: Three Key Trade-Offs

Zvi10 Sep 2019 12:10 UTC
47 points
4 comments5 min readLW link
(thezvi.wordpress.com)

Hack­able Re­wards as a Safety Valve?

Davidmanheim10 Sep 2019 10:33 UTC
18 points
17 comments1 min readLW link

Coun­ter­fac­tual Or­a­cles = on­line su­per­vised learn­ing with ran­dom se­lec­tion of train­ing episodes

Wei Dai10 Sep 2019 8:29 UTC
52 points
26 comments3 min readLW link

Is my re­sult wrong? Maths vs in­tu­ition vs evolu­tion in learn­ing hu­man preferences

Stuart_Armstrong10 Sep 2019 0:46 UTC
17 points
11 comments3 min readLW link

Shal­low Re­view of Con­sis­tency in State­ment Evaluation

Elizabeth9 Sep 2019 23:21 UTC
65 points
6 comments9 min readLW link

Sim­ple and com­pos­ite par­tial preferences

Stuart_Armstrong9 Sep 2019 23:07 UTC
11 points
0 comments2 min readLW link

Slat­estar­codex Meetup in Ham­burg, Germany

Gunnar_Zarncke9 Sep 2019 21:42 UTC
8 points
0 comments1 min readLW link

[Question] What tools ex­ist to com­pute all pos­si­ble pro­grams?

Alexei9 Sep 2019 16:50 UTC
16 points
2 comments1 min readLW link

[Question] Alter­na­tive name for a Bug List

Bae's Theorem9 Sep 2019 15:14 UTC
8 points
2 comments1 min readLW link

[Question] What are some pod­casts that just read aloud worth­while con­tent?

Rudi C9 Sep 2019 8:41 UTC
6 points
3 comments1 min readLW link

Sur­vival and Flour­ish­ing Fund grant ap­pli­ca­tions open un­til Oc­to­ber 4th ($1MM-$2MM planned for dis­per­sal)

habryka9 Sep 2019 4:12 UTC
27 points
0 comments1 min readLW link

[Question] Look­ing for an­swers about quan­tum im­mor­tal­ity.

Fivehundred9 Sep 2019 2:16 UTC
15 points
57 comments3 min readLW link

Sto­ries About Progress

Ben Pace8 Sep 2019 23:07 UTC
28 points
1 comment19 min readLW link

The Power to Make Scien­tific Breakthroughs

Liron8 Sep 2019 4:14 UTC
22 points
2 comments6 min readLW link

[Question] Do you have al­gorithms for pass­ing time pro­duc­tively with only your own mind?

Rudi C7 Sep 2019 20:48 UTC
8 points
4 comments1 min readLW link

Age gaps and Birth or­der: Reanalysis

Bucky7 Sep 2019 19:33 UTC
48 points
4 comments9 min readLW link

Age gaps and Birth or­der: Failed re­pro­duc­tion of results

Bucky7 Sep 2019 19:22 UTC
59 points
1 comment5 min readLW link

Are min­i­mal cir­cuits de­cep­tive?

evhub7 Sep 2019 18:11 UTC
78 points
11 comments8 min readLW link

Are we liv­ing at the most in­fluen­tial time in his­tory?

jp7 Sep 2019 18:09 UTC
14 points
0 comments1 min readLW link
(forum.effectivealtruism.org)

In­te­grat­ing the Lindy Effect

lsusr7 Sep 2019 17:38 UTC
53 points
11 comments2 min readLW link1 review

Lis­bon SSC Meetup #2

tamkin&popkin7 Sep 2019 13:17 UTC
1 point
0 comments1 min readLW link

Rus­sian x-risks newslet­ter, sum­mer 2019

avturchin7 Sep 2019 9:50 UTC
39 points
5 comments4 min readLW link

AI Safety “Suc­cess Sto­ries”

Wei Dai7 Sep 2019 2:54 UTC
125 points
27 comments4 min readLW link1 review

[Question] What are con­crete ex­am­ples of po­ten­tial “lock-in” in AI re­search?

Grue_Slinky6 Sep 2019 23:39 UTC
19 points
6 comments3 min readLW link

Con­crete ex­per­i­ments in in­ner alignment

evhub6 Sep 2019 22:16 UTC
74 points
12 comments6 min readLW link

Run­ning Effec­tive Struc­tured Fore­cast­ing Sessions

6 Sep 2019 21:30 UTC
21 points
0 comments3 min readLW link

Poli­ti­cal Violence and Dis­trac­tion Theories

Ben Pace6 Sep 2019 20:21 UTC
17 points
0 comments34 min readLW link

Copen­hagen SSC/​LW Meetup

Søren Elverlin6 Sep 2019 19:21 UTC
2 points
0 comments1 min readLW link

[Question] Does any­one else feel LessWrong is slow?

lifelonglearner6 Sep 2019 19:20 UTC
11 points
3 comments1 min readLW link

Ex­am­ples of Examples

Liron6 Sep 2019 14:04 UTC
19 points
7 comments1 min readLW link

Somerville Hous­ing Units

jefftk6 Sep 2019 14:00 UTC
7 points
0 comments2 min readLW link
(www.jefftk.com)

Who To Root For: 2019 Col­lege Foot­ball Edition

Zvi6 Sep 2019 8:10 UTC
11 points
2 comments43 min readLW link
(thezvi.wordpress.com)

SSC Mee­tups Every­where: Com­mu­nity Map on front­page this month

habryka5 Sep 2019 23:49 UTC
22 points
0 comments1 min readLW link

Seven habits to­wards highly effec­tive minds

Richard_Ngo5 Sep 2019 23:10 UTC
38 points
11 comments2 min readLW link
(thinkingcomplete.blogspot.com)

How to Throw Away Information

johnswentworth5 Sep 2019 21:10 UTC
18 points
5 comments5 min readLW link

[Question] How im­por­tant are model sizes to your timeline pre­dic­tions?

SoerenMind5 Sep 2019 17:34 UTC
11 points
1 comment1 min readLW link

Utility ≠ Reward

Vlad Mikulik5 Sep 2019 17:28 UTC
130 points
24 comments1 min readLW link2 reviews

Log­i­cal Coun­ter­fac­tu­als and Propo­si­tion graphs, Part 3

Donald Hobson5 Sep 2019 15:03 UTC
6 points
0 comments4 min readLW link

Somerville Hous­ing Over Time

jefftk5 Sep 2019 14:40 UTC
8 points
0 comments1 min readLW link
(www.jefftk.com)

Liv­ing the Berkeley idealism

Yuxi_Liu4 Sep 2019 22:20 UTC
12 points
3 comments1 min readLW link

Sto­ries About Education

Ben Pace4 Sep 2019 19:53 UTC
38 points
1 comment13 min readLW link

Kansas City SSC meetup

Bae's Theorem4 Sep 2019 15:27 UTC
4 points
0 comments1 min readLW link

The Power to Judge Startup Ideas

Liron4 Sep 2019 15:07 UTC
94 points
28 comments8 min readLW link

SSC Paris meetup

fbreton4 Sep 2019 9:33 UTC
5 points
0 comments1 min readLW link

Cach­ing on Suc­cess

pcm504 Sep 2019 6:34 UTC
14 points
1 comment2 min readLW link

Coun­ter­fac­tu­als are an An­swer, Not a Question

Chris_Leong3 Sep 2019 15:36 UTC
14 points
6 comments4 min readLW link