RSS

capybaralet

Karma: 215 (LW), 0 (AF)

X-risks are a tragedies of the commons

capybaralet
7 Feb 2019 2:48 UTC
9 points
19 comments1 min readLW link

My use of the phrase “Su­per-Hu­man Feed­back”

capybaralet
6 Feb 2019 19:11 UTC
12 points
0 comments1 min readLW link

Thoughts on Ben Garfinkel’s “How sure are we about this AI stuff?”

capybaralet
6 Feb 2019 19:09 UTC
25 points
17 comments1 min readLW link

The role of epistemic vs. aleatory un­cer­tainty in quan­tify­ing AI-Xrisk

capybaralet
31 Jan 2019 6:13 UTC
14 points
6 comments2 min readLW link

Imi­ta­tion learn­ing con­sid­ered un­safe?

capybaralet
6 Jan 2019 15:48 UTC
9 points
11 comments1 min readLW link

Con­cep­tual Anal­y­sis for AI Align­ment

capybaralet
30 Dec 2018 0:46 UTC
26 points
2 comments2 min readLW link

Disam­biguat­ing “al­ign­ment” and re­lated no­tions

capybaralet
5 Jun 2018 15:35 UTC
43 points
21 comments2 min readLW link

Prob­lems with learn­ing val­ues from observation

capybaralet
21 Sep 2016 0:40 UTC
0 points
4 comments1 min readLW link

Risks from Ap­prox­i­mate Value Learning

capybaralet
27 Aug 2016 19:34 UTC
1 point
10 comments1 min readLW link

Ineffi­cient Games

capybaralet
23 Aug 2016 17:47 UTC
14 points
13 comments1 min readLW link

Should we en­able pub­lic bind­ing pre­com­mit­ments?

capybaralet
31 Jul 2016 19:47 UTC
0 points
19 comments1 min readLW link

A Ba­sic Prob­lem of Ethics: Panpsy­chism?

capybaralet
27 Jan 2015 6:27 UTC
−4 points
16 comments1 min readLW link

A Some­what Vague Pro­posal for Ground­ing Ethics in Physics

capybaralet
27 Jan 2015 5:45 UTC
−3 points
15 comments1 min readLW link