A new mathematical framework called "differential homology," which allows us to count "holes" in shapes not just topologically, but also with calculus, considering their smoothness. This new tool could offer fresh ways to understand how continuous physical fields relate to the underlying topological structure of our universe.
Differential Homology · Fabio Ferrari Ruffino, Gabriel Longatto Clemente · arXiv:2608.05048
A new, robust mathematical tool to quantify how multiple sets of data points, like different cell types in a tissue, spatially interact by measuring a simple shape characteristic (like 'pieces minus holes') of their overlapping 'fuzzy' regions across various scales.
The Intersection Euler Characteristic Profile: Euler Calculus and Stability for Topological Interaction of Ball Unions · Kazuhiro Kawamura, Sushovan Majhi, Atish Mitra · arXiv:2608.06180
A new method to uncover the hidden rules governing a system, even when we only observe a small, incomplete part of it. It does this by treating seemingly random or irreversible processes as predictable movements in a larger, unseen reality.
Characteristic Sensitivity Ensembles for Inference of Hidden Dynamics from Marginal Observations · Qi Wang, Gustaaf Jacobs · arXiv:2608.06190
This paper establishes a fundamental mathematical rule for when advanced prediction tools, used in machine learning and understanding complex systems, will produce reliable and "well-behaved" results.
Verifiable Regularity Criterion for Conditional Expectation Operators and Conditional Mean Embeddings with Applications to Nonparametric Regression, Bayesian Inverse Problems, and Koopman Operators · Maximiliano Hertel, Ilja Klebanov, Manuel Schaller et al. · arXiv:2608.06155
A look at how a group of people, whose social connections constantly reshuffle, eventually all agree on one of two opinions. Even a slight initial advantage for one side can quickly lead to widespread agreement, precisely predicting the likelihood and speed of this consensus forming.
Majority Dynamics on Resampled Sparse Erdős--Rényi Graphs: Gaussian Winner Selection and Pace to Unanimity · Ioana Dumitriu, Muchen Ju, Hai-Xiao Wang · arXiv:2608.06159
This paper studies systems where things spread and multiply, like a disease or a rumor, and pinpoints exactly which individual elements are responsible for defining the leading edge. Only those elements whose paths stay incredibly close to the absolute front-runner actually contribute to pushing the overall "front" forward.
Almost sure path localisation for the derivative martingale of branching Brownian motion · Berestycki Julien, Chataigner Louis, Flath Gabriel · arXiv:2608.06263
A fresh perspective on knot theory, showing how complex properties of tangled loops can be understood through simple counting problems. Instead of using advanced algebra, it connects knot characteristics to elegant, organized patterns like "noncrossing partitions" and "cluster complexes."
Noncrossing Combinatorics, the Full Twist, and Decategorification of Knot Invariants · Colin Defant, Nathan Williams · arXiv:2608.06225