Alex_Altair

• Alex_AltairApr 12, 2025, 4:07 PM

  7 points

  4

  on: The In­ter­nal Model Prin­ci­ple: A Straight­for­ward Ex­pla­na­tion

  we only label states as ‘different’ if they actually result in different controller behaviour at some point down the line.

  This reminds me a lot of the coarse-graining of “causal” states in comp mech.

• Alex_AltairApr 4, 2025, 9:08 PM

  6 points

  0

  on: An­nounc­ing ILIAD2: ODYSSEY

  I got a ton of value from ILIAD last year, and strongly recommend it to anyone interested!

• Alex_AltairApr 4, 2025, 8:47 PM

  2 points

  0

  in reply to: Aella’s comment on: Con­sider showering

  IYKYK

Re­port & ret­ro­spec­tive on the Dove­tail fellowship

• Alex_AltairFeb 18, 2025, 1:50 AM

  4 points

  0

  in reply to: Alex_Altair’s comment on: Towards a for­mal­iza­tion of the agent struc­ture problem

  For anyone reading this comment thread in the future, Dalcy wrote an amazing explainer for this paper here.

Come join Dove­tail’s agent foun­da­tions fel­low­ship talks & discussion

Dove­tail’s agent foun­da­tions fel­low­ship talks & discussion

Towards build­ing blocks of ontologies

• Alex_AltairJan 29, 2025, 3:39 AM

  4 points

  0

  in reply to: quila’s comment on: An­nounce­ment: Learn­ing The­ory On­line Course

  See also the classic LW post, The Best Textbooks on Every Subject.

• Alex_AltairJan 15, 2025, 1:12 AM

  2 points

  0

  in reply to: Alexander Gietelink Oldenziel’s comment on: Alex_Al­tair’s Shortform

  Indeed, we know about those posts! Lmk if you have a recommendation for a better textbook-level treatment of any of it (modern papers etc). So far the grey book feels pretty standard in terms of pedagogical quality.

• Alex_AltairJan 14, 2025, 6:08 PM

  4 points

  0

  on: Alex_Al­tair’s Shortform

  Has anyone started accumulating errata for the SLT grey book? (I.e. Algebraic Geometry and Statistical Learning Theory by Sumio Watanabe.) This page on Watanabe’s website seems to just be about the Japanese version of the book.

• Alex_AltairJan 9, 2025, 9:19 PM

  LW: 4 AF: 2

  0

  AF

  on: Shal­low re­view of tech­ni­cal AI safety, 2024

  Some small corrections/​additions to my section (“Altair agent foundations”). I’m currently calling it “Dovetail research”. That’s not publicly written anywhere yet, but if it were listed as that here, it might help people who are searching for it later this year.

  Which orthodox alignment problems could it help with?: 9. Humans cannot be first-class parties to a superintelligent value handshake

  I wouldn’t put number 9. Not intended to “solve” most of these problems, but is intended to help make progress on understanding the nature of the problems through formalization, so that they can be avoided or postponed, or more effectively solved by other research agenda.

  Target case: worst-case

  definitely not worst-case, more like pessimistic-case

  Some names: Alex Altair, Alfred Harwood, Daniel C, Dalcy K

  Add “José Pedro Faustino”

  Estimated # FTEs: 1-10

  I’d call it 2, averaged throughout 2024.

  Some outputs in 2024: mostly exposition but it’s early days

  Basically right; I’d add this post and this post.

• Alex_AltairDec 30, 2024, 11:05 PM

  1 point

  0

  in reply to: eukaryote’s comment on: Learn to write well BEFORE you have some­thing worth saying

  “Gain writing skills BEFORE...”

• Alex_AltairDec 30, 2024, 10:14 PM

  3 points

  1

  in reply to: Kabir Kumar’s comment on: Shal­low re­view of live agen­das in al­ign­ment & safety

  FWIW I can’t really tell what this website is supposed to be/​do by looking at the landing page and menu

• Alex_AltairDec 30, 2024, 10:09 PM

  3 points

  −5

  on: Learn to write well BEFORE you have some­thing worth saying

  The title reads ambiguous to me; I can’t tell if you mean “learn to [write well] before” or “learn to write [well before]”.

• Alex_AltairDec 23, 2024, 4:19 AM

  5 points

  0

  on: Hire (or be­come) a Think­ing As­sis­tant /​ Body Double

  DM me if you’re interested.

  I, too am quite interested in trialing more people for roles on this spectrum.

• Alex_AltairDec 20, 2024, 9:13 PM

  3 points

  0

  in reply to: Zach Stein-Perlman’s comment on: o3

  Thanks. Is “pass@1” some kind of lingo? (It seems like an ungoogleable term.)

• Alex_AltairDec 20, 2024, 8:50 PM

  4 points

  0

  in reply to: Jonas Hallgren’s comment on: o3

  I guess one thing I want to know is like… how exactly does the scoring work? I can imagine something like, they ran the model a zillion times on each question, and if any one of the answers was right, that got counted in the light blue bar. Something that plainly silly probably isn’t what happened, but it could be something similar.

  If it actually just submitted one answer to each question and got a quarter of them right, then I think it doesn’t particularly matter to me how much compute it used.

• Alex_AltairDec 20, 2024, 7:37 PM

  7 points

  6

  in reply to: Alex_Altair’s comment on: o3

  On the livestream, Mark Chen says the 25.2% was achieved “in aggressive test-time settings”. Does that just mean more compute?

• Alex_AltairDec 20, 2024, 7:35 PM

  10 points

  1

  on: o3

  I wish they would tell us what the dark vs light blue means. Specifically, for the FrontierMath benchmark, the dark blue looks like it’s around 8% (rather than the light blue at 25.2%). Which like, I dunno, maybe this is nit picking, but 25% on FrontierMath seems like a BIG deal, and I’d like to know how much to be updating my beliefs.