--- Log opened Wed Feb 15 00:00:44 2023
01:02 -!- Llamamoe [~Llamamoe@46.204.76.113.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
01:08 -!- Llamamoe [~Llamamoe@46.204.76.113.nat.umts.dynamic.t-mobile.pl] has quit [Quit: Leaving.]
02:11 -!- Llamamoe [~Llamamoe@46.204.76.113.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
03:37 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
03:37 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
03:58 -!- srk [~sorki@user/srk] has quit [Quit: ZNC 1.8.1 - https://znc.in]
03:58 -!- srk [~sorki@user/srk] has joined #hplusroadmap
04:45 -!- Mabel [~Malvolio@idlerpg/player/Malvolio] has quit [Read error: No route to host]
04:52 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:20de:b029:6bab:f9d1] has joined #hplusroadmap
05:00 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
07:09 < kanzure> https://en.wikipedia.org/wiki/Embryonic_differentiation_waves
07:15 < kanzure> "Programming self-organizing multicellular structures with synthetic cell-cell signaling" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6492944/ (2018)
08:32 -!- Hooloovoo [~Hooloovoo@hax0rbana.org] has quit [Ping timeout: 248 seconds]
08:32 -!- Hooloovoo [~Hooloovoo@hax0rbana.org] has joined #hplusroadmap
08:33 -!- Malvolio is now known as Mabel
09:59 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has joined #hplusroadmap
10:22 < kanzure> the panda breeding people have a good incentive to fund infinity embryo splitting research
11:25 < yashgaroth> not much good for preserving genetic diversity, but good enough to fill zoos. I don't think pandas' situation is as dire as eg cheetahs on the genetic diversity front though
11:28 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has quit [Read error: Connection reset by peer]
13:43 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
13:43 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
15:32 -!- Llamamoe [~Llamamoe@46.204.76.113.nat.umts.dynamic.t-mobile.pl] has quit [Quit: Leaving.]
16:00 -!- nsh- is now known as nsh
16:44 -!- A_Dragon [A_D@libera/staff/dragon] has quit [Ping timeout: 620 seconds]
16:55 -!- A_Dragon [A_D@libera/staff/dragon] has joined #hplusroadmap
16:55 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Read error: Connection reset by peer]
17:14 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
17:22 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
17:38 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
17:38 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
17:39 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
17:41 < fenn> syntience had a pretty compelling tech demo in 2011 cramming dirty real world data into a symbolic layer (good old fashioned AI)
17:42 < fenn> it's possible the algorithms are actually better than state of the art, but due to lack of hardware scaling it's not getting as good results as more highly funded operations
17:43 < fenn> it doesn't use a GPU and there hasn't really been a lot of progress in CPUs over the past decade
17:43 < muurkha> yeah, and Monica's been through two AI winters already; this is one of the reasons I thought her post was worth pasting in here
17:44 < muurkha> some random person saying ChatGPT is the greatest invention since fire and qualifies as AGI, well, whatever
17:45 < muurkha> an AI researcher who I know from Hackers, it's a different situation
17:49 -!- Mabel is now known as ANACHRON
17:50 < fenn> what's "Hackers"?
17:53 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1075595646150651914/hqdefault.png
17:55 < fenn> i know the movie. i'll just pretend you gave an answer that made sense, i don't really care that much
18:00 < fenn> https://artificial-intuition.com/anderson.html she started in 1988 apparently
18:03 < fenn> i like the footnotes on the side style, that should be the default for web articles
18:03 < fenn> marginalia
18:13 < kanzure> https://en.wikipedia.org/wiki/Familial_natural_short_sleep
18:13 < kanzure> looks like a few different mutations have been added, along with some nice side effects
18:18 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:20de:b029:6bab:f9d1] has quit [Quit: Leaving]
18:22 < muurkha> fenn: the conference ;)
18:23 < muurkha> I thought she went back further than that
18:24 < muurkha> yeah, she does
18:44 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
18:44 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
19:48 < hprmbridge> kanzure> https://spectech.notion.site/Nanomodular-Electronics-7b78b981ae0c45eaaacc8cf778bdf751
20:14 -!- masamune [~masamune@47.154.199.139] has joined #hplusroadmap
20:29 < lsneff> just had a conversation with someone in which he called my assertion that the brain was effectively where the self is stored a “17th century white man’s idea of a person”
20:30 < lsneff> I think they’re super smart, but maybe a little too culture-brained
20:34 < masamune> where do you think the self is stored? - not that I agree with them, I am just curious
20:36 < muurkha> the brain, apparently?
20:36 < masamune> i misread that smh
20:37 < lsneff> Certainly within however persistent state is stored within the brain
20:37 < lsneff> Synapses don’t really have a ‘weight’, but that’s a simplified way of looking at it
20:38 < masamune> so if I downloaded my self to a computer, would I weigh less?
20:39 < lsneff> Sorry, weight in the neural network sense.
20:40 < masamune> I mean its energy, light, so its gotta have weight
20:40 < masamune> allegedly
20:40 < masamune> unless "self" is a percieved phenomena and not an actual "thing"
20:41 < lsneff> Well, that’s a complicated question
20:42 < lsneff> I didn’t use the term self when talking to them, I said something along the lines of personality and memories
20:45 < masamune> which would be more along the lines of self being a perceived thing - I wonder if you can calculate the physical weight of a memory
20:49 < lsneff> We don’t yet know all the physical changes that result from learning, so not at the moment
20:51 < muurkha> a common technique in analog electronics is to use differential signaling to get noise immunity: instead of transmitting a signal as a voltage on one wire, you transmit it as a difference of voltages on two wires
20:53 < muurkha> if you store data in such a differential form, you get similar advantages; aside from 8b/10b, a common quotidian example of this is constant-weight encoding for barcodes: https://en.wikipedia.org/wiki/Interleaved_2_of_5
20:56 < lsneff> Yeah, there’s no doubt that information is stored in a highly redundant manner
20:56 < lsneff> Differential data storage is interesting, hadn’t heard of that before
20:56 < lsneff> Makes sense for systematic I/o error
20:57 < muurkha> right, if you have some blooming of white or black areas, or a brightness gradient across the barcode, it allows you to cancel it out
20:57 < muurkha> so it wouldn't be very surprising if brains' memories were encoded in an approximately constant-weight form
21:02 < lsneff> What does constant-weight mean here?
21:02 < masamune> ^
21:04 < muurkha> well, a constant-weight binary code is one where all the codewords have the same Hamming weight, the same number of ones
21:05 < muurkha> like the 2-of-5 I linked above, where the 10 valid codewords are
21:05 < muurkha> 11000 10100 10010 10001 01100  01010 01001 00110 00101 00011
21:05 < muurkha> (or equivalently their complements)
21:06 < masamune> right 3 ones instead of 2
21:07 < muurkha> I'm vaguely gesturing at a generalization of this idea, where every possible thing you could encode is encoded as some ensemble of quantities that has the same sum as the encoding of every other thing you could encode
21:08 < muurkha> maybe you increase the coupling factors of 2000 synapses and decrease the coupling factors of 2000 others, for example
21:08 < masamune> clever
21:09 < muurkha> that way if the brain is operating at a higher serotonin level or something, it doesn't activate any particular set of memories
21:09 < muurkha> so it might be that a memory has no weight at all, any more than a hard disk using 8b/10b weighs more after you store data on it
21:10 < muurkha> https://en.wikipedia.org/wiki/8b/10b_encoding
21:10 < muurkha> 8b/10b is not *precisely* constant weight, but it's pretty close
21:10 < masamune> kind of like a hologram
21:11 < muurkha> yeah, holograms are another example of this approach
21:11 < muurkha> and hyperdimensional computing is explicitly inspired by holography
21:12 < masamune> in this model the total weight of the system doesn't change, but the encoded information within the system has a weight within that system
21:13 < muurkha> in a sense, sure, but it's the same weight as any other possible encoded information
21:23 -!- masamune [~masamune@47.154.199.139] has quit [Read error: Connection reset by peer]
21:43 -!- ANACHRON is now known as Mabel
22:04 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 255 seconds]
--- Log closed Thu Feb 16 00:00:45 2023