Learning 3-manifold triangulations
In this article, the authors evalute the performance of modern machine learning tools on the problem of recognizing 3-manifolds and knot complements. The starting point of their analysis is to choose a small collection of manifolds of interest and generate a large number of triangulations of these manifolds using Pachner moves. They first provide an analysis of the obtained Pachner graphs from the perspective of network science: centrality, Wiener index, clustering coefficients, etc., leading them to an intriguing conjecture connecting the growth rate of the Pachner graph of a cusped hyperblic 3-manifold to the inverse of the systole.\N\NThen, these triangulations are used to train and evaluate the performance of neural networks at recognizing 3-manifolds from their triangulations, and more precisely from the isomorphism signatures of their triangulations, which are very concise string encodings of their gluing structures. Extensive results are provided, highlighting in particular that some pairs of manifolds are harder to distinguish than others, and that the transformer architecture, despite being notably well-tailored to string problems, does not perform well on the task at hand.
- Advanced Lectures on Machine Learning
- Advancing mathematics by guiding human intuition with AI
- Algorithmic homeomorphism of 3-manifolds as a corollary of geometrization
- Algorithmic topology and classification of 3-manifolds
- An Imbedding Theorem for Connected 3-Manifolds with Boundary
- Brain webs for brane webs
- Branes with brains: exploring string vacua with deep reinforcement learning
- Classes of cycle bases
- Cluster algebras: network science and machine learning
- Collective dynamics of `small-world' networks
- Constructing and machine learning Calabi-Yau five-folds
- Deep learning the hyperbolic volume of a knot
- Die Gruppe der Abbildungsklassen. (Das arithmetische Feld auf Flächen.)
- Evolving Heterotic Gauge Backgrounds: Genetic Algorithms versus Reinforcement Learning
- Evolving neural networks with genetic algorithms to study the string landscape
- Hilbert series, machine learning, and applications to physics
- scientific article; zbMATH DE number 4100181 (Why is no real title available?)
- scientific article; zbMATH DE number 1467264 (Why is no real title available?)
- scientific article; zbMATH DE number 849995 (Why is no real title available?)
- Knots are determined by their complements
- Learning knot invariants across dimensions
- Machine Learning Clifford invariants of ADE Coxeter elements
- Machine learning in the string landscape
- Machine Learning Line Bundle Cohomology
- Machine learning Sasakian and G₂ topology on contact Calabi-Yau 7-manifolds
- Machine-learning dessins d’enfants: explorations via modular and Seiberg–Witten curves
- Neurons on amoebae
- New Calabi-Yau manifolds from genetic algorithms
- Physical Yukawa couplings in heterotic string compactifications
- Quantum field theory and the Jones polynomial
- Scikit-learn: machine learning in Python
- Simplifying triangulations of \(S^3\).
- STRUCTURES OF SMALL CLOSED NON-ORIENTABLE 3-MANIFOLD TRIANGULATIONS
- The homeomorphism problem for closed 3-manifolds
- The Pachner graph and the simplification of 3-sphere triangulations
- Three dimensional manifolds, Kleinian groups and hyperbolic geometry
- Topology and physics
- Topology for Computing
- TRANSFORMATIONS OF SPECIAL SPINES AND THE ZEEMAN CONJECTURE
- Zur Theorie der Matrices.
- Über Matrizen aus nicht negativen Elementen.
This page was built for publication: Learning 3-manifold triangulations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7004128)
