Anything related to chaos theory, network science, complex systems, emergent phenomena, or whatever you want to call this large class of things and ideas, it's going to be heavy, like population level mind control psychometrics heavy.
From branches to loops: The physics of transport networks in nature
Sep 2024, phys.org
I have noticed that when papers from Poland make it to the main feed you know you're in for some shit:
An important advantage of looping networks is their reduced vulnerability to damage; in networks without loops, the destruction of one branch can cut off all connected branches, whereas in networks with loops, there is always another connection to the rest of the system.Many transport networks grow in response to a diffusive field, such as the concentration of a substance, the pressure in the system, or the electric potential. The fluxes of such a field are much more easily transported through the branches of the network than through the surrounding medium."We showed that a small difference in resistance between the network and the medium can lead to attraction between growing branches and the formation of loops.""Analyzing the development of these [jellyfish gastrovascular] canals over time, I noticed that when one of them connects to the jellyfish's stomach (the boundary of the system) then the shorter canals are immediately attracted to it and form loops.""Our model predicts that the attraction between neighboring branches after a breakthrough occurs regardless of the geometry of the network or the difference in resistance between the network and the surrounding medium." -Prof. Piotr Szymczak from the Faculty of Physics at the University of Warsaw
via University of Warsaw: Stanisław Żukowski et al, Breakthrough-induced loop formation in evolving transport networks, Proceedings of the National Academy of Sciences (2024). DOI: 10.1073/pnas.2401200121
Image credit: AI Art - 3D Software Code Structure - 2025
Computational method pinpoints how cause-and-effect relationships ebb and flow over time
Nov 2024, phys.org
"Currently available methods for studying complex systems tend to assume that the system is approximately stationary—that is, the system's dynamical properties stay the same over time. Other commonly introduced simplifications such as linearity and time invariance can produce incorrect expectations that fail to quantify changes in the strength or direction of these relationships."To address this gap, the research team developed a novel machine-learning model called Temporal Autoencoders for Causal Inference (TACI) to identify and measure the direction and strength of causal interactions that vary over time.They used an established model of a dynamic system but generated a dataset where interactions (couplings) changed over time, and found that TACI was able to detect how the strength of the causal relationship changed.Next they looked at real data, starting with weather data, and found that causal interactions peak during times when the temperature drops—demonstrating that TACI can accurately predict true variations over time from messy real-world data. Then with brain data on anesthetized monkeys, and found almost all interactions disappear during the anesthetized period, and then begin to re-emerge during recovery.
via Department of Physics Emory University: Josuan Calderon et al, Inferring the time-varying coupling of dynamical systems with temporal convolutional autoencoders, eLife (2024). DOI: 10.7554/eLife.100692.1
No comments:
Post a Comment