banner
elmarbucher.bsky.social
@elmarbucher.bsky.social
🇺🇦
P(X,H,O)
P: joint probability.
X: external observation.
H: internal hidden state.
O: output.
o = f(x,h)
+ https://orcid.org/0000-0002-2929-2460
+ https://github.com/elmbeech
+ https://elmbeech.github.io
🦕🦈 we are IU ~ never daunted!
Just finished reading @blaiseaguera.bsky.social 's book “What Is Intelligence?”
It took me a while. However, it is an amazing read! All 3 pats deserve to be read, although the first 2 parts I enjoyed most.
Above all, this book totally changed my view on AI (especially AGI)! Big fan.
December 20, 2025 at 11:27 PM
a while ago i was programming very little in Julia, which looks like a fun language too. but in my field, it never has this sound ecosystem of libraries that Python3 offers.
November 17, 2025 at 1:19 AM
i'm sorry to hear that you feel that way about python. the last 3 years (time flies) i was coding a lot in C and C++.
for us, the ones who learn programming back in the eighties, Python was (and i think still is) that modern language you talked about.
November 17, 2025 at 1:11 AM
i feel the same way! it's a full fledged language.
November 14, 2025 at 5:26 AM
Lana, what is your favourite programming language? I am not judging. I am just interested to hear. what is the language you get your daily work done?
November 14, 2025 at 5:09 AM
i have to think more about the separating logic from the logistic argument. this is a good discussion we have here.
look, in the end, it is a language. it is the feeling when you write in this language. And this can fall on both sides. i love to code in Python3.
November 14, 2025 at 4:59 AM
i read your post. i think it is interesting what you say! for me R is just Pandas. and as you said, the examples were very similar. for e.g. single cell analysis, i prefer e.g. scanpy and squidpy over seurat. manipulating plots, is maybe not that easy, as you highlighted.
November 14, 2025 at 4:49 AM
I learned my computer skills back in the days with Basic (C=64) and Pascal. And Python is so much closer to these languages than R.
I like to read your post, but i will not go back to R :" ).
November 14, 2025 at 4:03 AM
i worked in bioinformatics (and now mathematical biology) for more than 21 years. i worked with Perl, R / bioconductor, and Python3, and i can't help, from all these languages i like Python3 the most.
November 14, 2025 at 4:02 AM
i am not yet finished reading... however, what i read so far, this book is amazing! i highly recommend reading it. i think Blaise Agüera y Arcas is here onto something. a deeper, profound understanding of what really goes on. life is computation. excellent!
October 4, 2025 at 5:50 PM
@msahsorin.bsky.social this is so cool work!
September 21, 2025 at 1:52 AM
Ilya Prigogine's legacy :").
bsky.app/profile/norm... and extropic ai seem to be the ones who work on this technology. definitely a technology to keep an eye on!
bsky.app
August 16, 2025 at 10:49 PM
this is interesting! yesterday i stumbled over this. a thermodynamics computing accelerator. maybe in short term more feasible than a quantum computer accelerator? one will definitely not replace the other.

+ www.nature.com/articles/s41...
Thermodynamic computing system for AI applications - Nature Communications
Current digital hardware struggles with high computational demands in applications such as probabilistic AI. Here, authors present a small-scale thermodynamic computer composed of eight RLC circuits, ...
www.nature.com
August 16, 2025 at 9:43 PM