Constructive Bounds for the Topological Expressivity of Neural Networks
math.AT, cs.LG
Submitted: 2026-10-09
Updated: 2026-10-09
Sources
- A Quotient Homology Theory of Representation in Neural Networks
- The Real Tropical Geometry of Neural Networks
- A Topology Layer for Machine Learning
- On transversality of bent hyperplane arrangements and the topological expressiveness of ReLU neural networks
- On Characterizing the Capacity of Neural Networks using Algebraic Topology
- Auto-Encoding Variational Bayes
- Fourier Neural Operator for Parametric Partial Differential Equations
- Algebra Unveils Deep Learning -- An Invitation to Neuroalgebraic Geometry
- Efficient Estimation of Word Representations in Vector Space
- Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity
- Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
- Low-dimensional topology of deep neural networks