Catastrophic Goodhart in RL with KL penalty

Thomas Kwa and Adrià Garriga-alonso

15 May 2024 0:58 UTC

31 points

0 comments7 min readLW link

Ilya Sutskever and Jan Leike resign from OpenAI

Zach Stein-Perlman15 May 2024 0:45 UTC

72 points

19 comments1 min readLW link

(www.nytimes.com)

my note system

bhauth15 May 2024 0:20 UTC

19 points

2 comments2 min readLW link

(www.bhauth.com)

Teaching CS During Take-Off

andrew carle14 May 2024 22:45 UTC

50 points

4 comments2 min readLW link

A Narrative History of Environmentalism’s Partisanship

Jeffrey Heninger14 May 2024 16:51 UTC

25 points

1 comment10 min readLW link

(blog.aiimpacts.org)

How much AI inference can we do?

Benjamin_Todd14 May 2024 15:10 UTC

13 points

4 comments5 min readLW link

(benjamintodd.substack.com)

How to do conceptual research: Case study interview with Caspar Oesterheld

Chi Nguyen14 May 2024 15:09 UTC

40 points

3 comments9 min readLW link

[Question] In the context of AI interp. What is a feature exactly?

f3mi14 May 2024 13:46 UTC

8 points

0 comments1 min readLW link

Can AI partners make human ones pale in comparison?

false14 May 2024 10:57 UTC

−1 points

0 comments5 min readLW link

(arminbagrat.com)

[Question] Why do we enjoy music?

metachirality14 May 2024 8:29 UTC

7 points

2 comments1 min readLW link

Emergence Series

James Stephen Brown14 May 2024 8:06 UTC

4 points

0 comments1 min readLW link

(nonzerosum.games)

D&D.Sci Long War: Defender of Data-mocracy Evaluation & Ruleset

aphyer14 May 2024 3:35 UTC

38 points

3 comments6 min readLW link

Environmentalism in the United States Is Unusually Partisan

Jeffrey Heninger13 May 2024 21:23 UTC

69 points

5 comments4 min readLW link

(blog.aiimpacts.org)

Against Student Debt Cancellation From All Sides of the Political Compass

Maxwell Tabarrok13 May 2024 14:55 UTC

39 points

16 comments5 min readLW link

(www.maximum-progress.com)

What you really mean when you claim to support “UBI for job automation”: Part 1

Deric Cheng13 May 2024 8:52 UTC

13 points

11 comments10 min readLW link

The two-tiered society

Roman Leventov13 May 2024 7:53 UTC

4 points

9 comments3 min readLW link

[Question] Benefitial habits/personal rules with very minimal tradeoffs?

Slapstick13 May 2024 6:06 UTC

16 points

12 comments1 min readLW link

Retrospective on Mathematical Boundaries Workshop

miyazono and Chipmonk

12 May 2024 21:58 UTC

21 points

0 comments4 min readLW link

(formalizingboundaries.substack.com)

Understanding rationality vs. ideology debates

Aryeh Englander12 May 2024 19:20 UTC

13 points

1 comment6 min readLW link

Partitioned Book Club

jenn12 May 2024 18:38 UTC

10 points

5 comments4 min readLW link