A Dynamical Microscope for Multivariate Oscillatory Signals: Validating Regime Recovery on Shared Manifolds Journal: NeuroImage

Łukasz Furman, Ludovico Minati, Włodzisław Duch · 2026

Read A Dynamical Microscope for Multivariate Oscillatory Signals: Validating Regime Recovery on Shared Manifolds Journal: NeuroImage by Łukasz Furman, Ludovico Minati, Włodzisław Duch (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-11054.txt.

Description

Signals from brains, bodies, and machines often switch regimes without clean breaks. Łukasz Furman and two co-authors build a dynamical microscope that turns multichannel oscillatory signals into circular phase-amplitude traits, then learns a latent path with an autoencoder. Regime shifts are read from path shape and flow, not from hard state labels.

Checks on a coupled oscillator network with switching topology shows the tool recovers law changes even when regimes overlap in state space. Speed and explored variance separate regimes strongly, while other metrics capture shape apart from topology. The frame suits neural, physiological, and physical time series where regime structure hides in motion, not in static clusters.

Catalog record

CREATORŁukasz Furman, Ludovico Minati, Włodzisław Duch
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-11054

Browse the full library · About the Tower of Babel