RSS

Stuart_Armstrong(Stuart Armstrong)

Karma: 22,744

Com­par­ing re­ward learn­ing/​re­ward tam­per­ing formalisms

Stuart_Armstrong
21 May 2020 12:03 UTC
9 points
0 comments3 min readLW link

Prob­a­bil­ities, weights, sums: pretty much the same for re­ward functions

Stuart_Armstrong
20 May 2020 15:19 UTC
11 points
1 comment2 min readLW link

Learn­ing and ma­nipu­lat­ing learning

Stuart_Armstrong
19 May 2020 13:02 UTC
38 points
4 comments10 min readLW link

Re­ward func­tions and up­dat­ing as­sump­tions can hide a mul­ti­tude of sins

Stuart_Armstrong
18 May 2020 15:18 UTC
16 points
2 comments9 min readLW link

How should AIs up­date a prior over hu­man prefer­ences?

Stuart_Armstrong
15 May 2020 13:14 UTC
17 points
9 comments2 min readLW link

Dist­in­guish­ing lo­gis­tic curves

Stuart_Armstrong
15 May 2020 11:38 UTC
23 points
0 comments7 min readLW link

Dist­in­guish­ing lo­gis­tic curves: visual

Stuart_Armstrong
15 May 2020 10:33 UTC
9 points
0 comments1 min readLW link

Kurzweil’s pre­dic­tions’ in­di­vi­d­ual scores

Stuart_Armstrong
7 May 2020 17:10 UTC
17 points
0 comments1 min readLW link

Assess­ing Kurzweil pre­dic­tions about 2019: the results

Stuart_Armstrong
6 May 2020 13:36 UTC
109 points
13 comments4 min readLW link

Maths writer/​cowrit­ter needed: how you can’t dis­t­in­guish early ex­po­nen­tial from early sigmoid

Stuart_Armstrong
6 May 2020 9:41 UTC
39 points
13 comments1 min readLW link

Con­sis­tent Glo­ma­riza­tion should be feasible

Stuart_Armstrong
4 May 2020 10:06 UTC
10 points
11 comments1 min readLW link

Last chance for as­sess­ing Kurzweil

Stuart_Armstrong
22 Apr 2020 11:51 UTC
12 points
0 comments1 min readLW link

Databases of hu­man be­havi­our and prefer­ences?

Stuart_Armstrong
21 Apr 2020 18:06 UTC
10 points
9 comments1 min readLW link

So­lar sys­tem colon­i­sa­tion might not be driven by economics

Stuart_Armstrong
21 Apr 2020 17:10 UTC
24 points
44 comments2 min readLW link

“How con­ser­va­tive” should the par­tial max­imisers be?

Stuart_Armstrong
13 Apr 2020 15:50 UTC
20 points
8 comments2 min readLW link

Assess­ing Kurzweil’s 1999 pre­dic­tions for 2019

Stuart_Armstrong
8 Apr 2020 14:27 UTC
37 points
9 comments1 min readLW link

Call for vol­un­teers: as­sess­ing Kurzweil, 2019

Stuart_Armstrong
2 Apr 2020 12:07 UTC
27 points
21 comments1 min readLW link

An­throp­ics over-sim­plified: it’s about pri­ors, not updates

Stuart_Armstrong
2 Mar 2020 13:45 UTC
9 points
0 comments1 min readLW link

If I were a well-in­ten­tioned AI… IV: Mesa-optimising

Stuart_Armstrong
2 Mar 2020 12:16 UTC
25 points
2 comments6 min readLW link

If I were a well-in­ten­tioned AI… III: Ex­tremal Goodhart

Stuart_Armstrong
28 Feb 2020 11:24 UTC
19 points
0 comments5 min readLW link