Neural Networks for Threshold Dynamics Reconstruction

Elisa Negrini, Almanzo Jiahe Gao, Abigail Bowering, Wei Zhu, Luca Capogna · 2024

Read Neural Networks for Threshold Dynamics Reconstruction by Elisa Negrini, Almanzo Jiahe Gao, Abigail Bowering, Wei Zhu, Luca Capogna (2024) 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-2412-09079.txt.

Description

Elisa Negrini and co-authors teach neural networks to imitate the marching of fronts: melting ice edges, spreading fire lines, and other boundaries that move by threshold rules. Their 2024 paper builds two convolutional architectures inspired by a classic front-propagation algorithm and by cellular automata, learning the dynamics straight from video.

The first network learns one specific dynamic per video, kernel and threshold fitted to that footage. The second, a meta-learning version, adapts its parameters to new dynamics. Both reconstructed and extrapolated evolving boundaries well on synthetic and real videos of ice melting and fire spread, with noise corrupting the input. Front evolution sits inside materials science, wildfire response, and image processing alike. A learned simulator that reads a front's habits from a clip could forecast where the line goes next, which is the game in each of those fields.

Catalog record

CREATORElisa Negrini, Almanzo Jiahe Gao, Abigail Bowering, Wei Zhu, Luca Capogna
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE75 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-09079

Browse the full library · About the Tower of Babel