Abstract:How can Transformers model and learn enumerative geometry? What is a robust procedure for using Transformers in abductive knowledge discovery within a mathematician-machine collaboration? In this work, we introduce a new paradigm in computational enumerative geometry in analyzing the $\psi$-class intersection numbers on the moduli space of curves. By formulating the enumerative problem as a continuous optimization task, we develop a Transformer-based model for computing $\psi$-class intersection numbers based on the underlying quantum Airy structure. For a finite range of genera, our model is capable of regressing intersection numbers that span an extremely wide range of values, from $10^{-45}$ to $10^{45}$. To provide a proper inductive bias for capturing the recursive behavior of intersection numbers, we propose a new activation function, Dynamic Range Activator (DRA). Moreover, given the severe heteroscedasticity of $\psi$-class intersections and the required precision, we quantify the uncertainty of the predictions using Conformal Prediction with a dynamic sliding window that is aware of the number of marked points. Next, we go beyond merely computing intersection numbers and explore the enumerative "world-model" of the Transformers. Through a series of causal inference and correlational interpretability analyses, we demonstrate that Transformers are actually modeling Virasoro constraints in a purely data-driven manner. Additionally, we provide evidence for the comprehension of several values appearing in the large genus asymptotic of $\psi$-class intersection numbers through abductive hypothesis testing.

Compression with wildcards: All induced metric subgraphs

Compression with wildcards: All induced subgraphs that are (respectively) connected, chordal, bipartite, or forests

Enumerating minimal solution sets for metric graph problems

Combinatorial invariants of metric filtrations and automorphisms; the universal adic graph

Beyond trees: the metric geometry of subsets of weighted Hamming cubes

Physical Schemata Underlying Biological Pattern Formation - Examples, Issues and Strategies

Generating cut conjunctions and bridge avoiding extensions in graphs

Metric dimensions of generalized Sierpiński graphs over squares

On the metric representation of the vertices of a graph

Enumerating all geodesics

Combinatorial invariants of finite metric spaces and the Wasserstein arrangement

Can Transformers Do Enumerative Geometry?

Metric thickenings of Euclidean submanifolds

Combinatorial characterization of pseudometrics

On Sharp Bounds of Local Fractional Metric Dimension for Certain Symmetrical Algebraic Structure Graphs

Metric Multiview Geometry -- a Catalogue in Low Dimensions

Optimal (Euclidean) Metric Compression

Variety of mutual-visibility problems in graphs

Towards a Calculus for Non-Linear Spectral Gaps [Extended Abstract]

Combinatorial degree bound for toric ideals of hypergraphs