@bnut Flocks on a box
I let the simulation run for several minutes before recording. The edges do appear to be a bit sticky causing the flocks have a bias to moving parallel to them, but particles are only briefly trapped there before being pulled away.
#AlgorithmicArt #CreativeCoding #simulation #swarming #ALife
#Processing #glsl #shaders
@bnut Swarms on a seam
The particles try to move in 3-D, but then their position is constrained so that z = abs(x).
Still seems to work with this particular set of forces, though it's not all that easy to see in this low-res, short animation.
#AlgorithmicArt #CreativeCoding #simulation #swarming #ALife
#Processing #glsl #shaders
After a long collaboration with @martinbiehl, @mc and @Nathaniel I’m excited to share the first of (hopefully) many outputs:
“A Bayesian Interpretation of the Internal Model Principle”
https://arxiv.org/abs/2503.00511.
This work combines ideas from control theory, applied #categorytheory and #Bayesian reasoning, with ramifications for #cognitive science, #AI/#ML, #ALife and biology to be further explored in the future.
In these fields, we come across ideas of “models”, “internal models”, “world models”, etc. but it is hard to find formal definitions, and when one does, they usually aren’t general enough to cover all the aspects these different fields consider important.
In this work, we focus on two specific definitions of models, and show their connections. One is inspired by work in control theory, and one comes from Bayesian inference/filtering for cognitive science, AI and ALife, and is formalised with Markov categories.
In the first part, we review and reformulate the “internal model principle” from control theory (at least, one of its versions) in a more modern language heavily inspired by categorical systems theory (https://www.davidjaz.com/Papers/DynamicalBook.pdf, https://github.com/mattecapu/categorical-systems-theory/blob/master/main.pdf).
Released #LiSE 0.17.0 https://github.com/TacticalMetaphysics/LiSE/releases/tag/v0.17.0
This release contains no new features, but changes some of the fundamentals of the data model to behave more reliably. It should be much more stable.
We have published a new edition of the Artificial Life newsletter! Check it out here:
https://alife-newsletter.github.io/Newsletter/edition_018.html
This time we feature three alife-y software projects contributed by our readers, a post-mortem of the ALIFE 2024 conference, two robots controlled by... weird stuff, and some cool art.
Please check it out, and consider contributing for the 19th edition, which we plan to release in December!
(You can also follow the newsletter by RSS, or join our mailing list)
Nils Aall Barricelli's pioneering work on #ALife simulations in the 1950s laid the foundation for artificial life research. His experiments simulated digital symbio-organisms as simple integer numbers. Made with #python, #matplotlib, #numpy, #numba.
the loops that form are very fragile. an open string will eventually break into little Ls. I need a cost function that discourages them from forming
instead of writing rules directly, I now have a cost function (energy) and saying a cell can move somewhere if the total energy (in a local region - up to 2 cells away) does not increase. this is fun because it can create some unexpected behaviour.
here the energy is defined as abs(2-num_neighbours), so having two neighbours is optimal. but they can join into pairs and still move around. the little L-shaped things are technically loops as all the cells have two neighbours, but I want bigger loops.
green cells can move randomly but must have two neighbours (and the neighbours must have two neighbours) = membrane
#cellularautomata #alife #creativecoding
New run of my "builders" #alife task. The bots run a small NN optimized to move "food" into clumps.
This variant gives higher reward when food is also next to a wall. NN weights were optimized with sep-CMA-ES.
Did some more brainstorming about a title for this project/piece and have settled on STRATA for now — these emerging structures/fragments very much remind me of them (rock strata and/or thin section microscopy) and the many folds & cross-stratifications observed on my hikes over the past few years have been a regular inspiration to keep working on this project...
Also see:
- https://mastodon.thi.ng/@toxi/111173324137924265
- https://mastodon.thi.ng/@toxi/109931051187422957
Been slacking posting more art here, so time for a teensy selection of an old generative/evolutionary system from 2014 (then used for my HOLO 2 magazine guest design). Originally written in Clojure, meanwhile ported to TypeScript & Zig, I've kept working on & experimenting with it ever since... 1000s of screenshots and 100s of versions to sift through. Loosely based on research done by Barricelli[1] since the early 1950s, conceptually and aesthetically it sits nicely between my C-SCAPE and De/Frag and has a similarly huge design space to explore (in some versions coupled with genetic programming to evolve cell replication rules)... There's a 1500 word draft blog post from back then too, which goes into more detail and history of this approach. Maybe its time to publish that one too at last... :)
My latest experiment involves evolving "interesting" Game of Life simulations. Some of the results are pretty neat, so I'll reply to this post to share videos as I come across ones I like...
#evolution #alife #gameoflife #gif
#genuary2023 Black & White. Some WIP snapshots of last year's C-SCAPE co-evolving 1D #CellularAutomata #GenerativeArt project...
Realtime versions:
https://art.thi.ng/c-scape
Co-evolving here means that there're multiple CAs sharing the same environment, each with a different neighborhood (shape & size), different rules and its own moving fuzzy mask/transition zone. Neighborhoods can also have a temporal dimension and as such are technically 1.5D or 2D and able to look up cell states from the (near) past. Where CAs overlap spatially, hybrid forms are emerging and these interactions can lead to entirely new outcomes...
The #OpenSource library which evolved from this project:
https://thi.ng/cellular
I got interested in #BiologicallyInspiredComputing when I learned about #ArtificialLife #ALife. At that time the computing resources available were limited compared to today. Now we have #DeepNeuralNetworks #DNN but it is widely agreed (including by me) that they do not replace natural #cognition. #BiologicallyInspiredComputing can be used as an application technology, but how do we use #Computing to understand #Cognition?
Hi, I'm Andrew!
Professionally, I'm a software engineer and engineering leader.
I slightly love front-end dev more than back-end, but in practice I'm an absolute generalist and really love seeing an entire architecture come together!
For fun, I might still be engulfed in #WebDev, but I also enjoy #3DPrinting, #GenerativeArt and nerding out about #ALife and robust-first computational stacks.
:
https://andrewwalpole.com
https://github.com/walpolea
https://codepen.io/walpolea
https://mfm.rocks
