--- Log opened Sun May 25 00:00:53 2025
00:55 -!- okelano [~okelano@m90-131-41-219.cust.tele2.lt] has quit [Ping timeout: 240 seconds]
00:58 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
02:38 < fenn> .t https://osf.io/preprints/psyarxiv/5g2xj_v3
02:38 < saxo> OSF
02:38 < fenn> "Ingressing Minds: Causal Patterns Beyond Genetics and Environment in Natural, Synthetic, and Hybrid Embodiments" by michael levin (yes that one)
02:45 < TMA> fleeky: that's discouraging. But it is understandable
03:59 -!- okelano [~okelano@m90-131-41-219.cust.tele2.lt] has joined #hplusroadmap
04:10 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 264 seconds]
04:12 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Quit: Avoid fossil fuels and animal products. Have no/fewer children. Protest, elect sane politicians. Invest ecologically.]
04:13 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
04:20 < L29Ah> 07:08:23]<fenn> https://dirtypcbs.com/about some kinda slowboat direct from china PCB and other stuffs manufacturing service related to hackaday
04:20 < L29Ah> seems more expensive than jlcpcb
04:22 < fenn> could be
04:23 < fenn> i was intrigued by the local china mail forwarding, with the stated intent that you can get chip samples from manufacturers and try them on a pcb
04:24 < fenn> under "other services"
04:36 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
04:36 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
05:34 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
06:06 -!- okelano [~okelano@m90-131-41-219.cust.tele2.lt] has quit [Ping timeout: 240 seconds]
06:20 -!- okelano [~okelano@m90-131-41-219.cust.tele2.lt] has joined #hplusroadmap
07:03 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Quit: Avoid fossil fuels and animal products. Have no/fewer children. Protest, elect sane politicians. Invest ecologically.]
07:03 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
09:10 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
09:10 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
09:12 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 264 seconds]
10:04 -!- okelano [~okelano@m90-131-41-219.cust.tele2.lt] has quit [Quit: Client closed]
10:15 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
10:21 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 264 seconds]
10:42 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
10:42 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Client Quit]
14:48 < hprmbridge> kanzure> https://www.ycombinator.com/launches/NUd-partcl-design-chips-in-minutes
14:50 < fleeky> fenn: https://7rnx.net/plato.jpg  (in regards to all that spicy levin research) 
14:52 < fleeky> had a long discussion with my partner about a lot of levin's ideas , not sure if we got quite to the platonic space of ideas but in general the stance they took was basically , intereting theory / approach but not actually reality .. personally i dont really know but the experimental results are why i take it seriously at all
15:25 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
15:28 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 260 seconds]
15:28 -!- stipa_ is now known as stipa
15:57 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Quit: Konversation terminated!]
17:01 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
17:14 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
17:23 < hprmbridge> alonzoc> Ye, I don't think it is "Platonic space", any attempt to compress/predict human text draws from very similar distributions (both generated by "reality") so the latent spaces of two models are going to have a lot of mutual information if they are modelling the same thing.
17:23 < hprmbridge> alonzoc> Then the easy to learn regularities are fairly universally easy to learn.
17:23 < hprmbridge> alonzoc> So you end up with latent spaces with less noise than reality which you already know have to be high in mutual information and then commonality of model architectures restricts the structure of the space.
17:28 < hprmbridge> alonzoc> The way our models work means that the embeddings of `man` `woman` `husband` and `wife` in any model are almost certainly to have the structure `husband - man + woman ~ wife`. And if two latent spaces have such a constraint on their structure you know almost certainly that there exists some invertable affine map from one to the other which preserves semantic meaning.
17:29 < hprmbridge> alonzoc> `s/almost certainly//2`
17:55 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Ping timeout: 265 seconds]
17:59 < fleeky> oggling random reciprocal framing concepts , this one looks almost like cheating : https://www.dfab.arch.ethz.ch/data/ProjectImages/02_Web/M/234/191030_234_FutureTree_Diagrams_Screw_Detail1_Aleksandra_Apolinarska_WM.jpg
18:28 < hprmbridge> Lev> ...I'm highly suspicious
18:29 < hprmbridge> Lev> Maybe the gate resizing and placement algs are Really Good™️ but I'm scared to see what'll come out if they "Apply AI" to functional reduction, etc
18:30 < hprmbridge> Lev> Also, if the placement algs are that much better than either the old algs were really poorly implemented or there's smth odd going on
18:31 < hprmbridge> nmz787> Particle sounds good, but it's not even layout from the sound
18:32 < hprmbridge> Lev> Last I was aware we already had PTAS's for placement problems since they're a euclidean case
18:33 < fenn> aka quarter isogrid
18:33 < fenn> ^fleeky
18:33 < hprmbridge> Lev> So if this manages to improve that by some 100 times either that's some great low-hanging fruit to reclaim elsewhere as well, or something's weird with NP
18:34 < hprmbridge> Lev> Or just poor benchmarking
19:23 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
19:24 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
21:06 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Quit: justanotheruser]
21:10 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
23:18 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
23:18 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
--- Log closed Mon May 26 00:00:53 2025