A unified framework linking compactness, definability, and learnability, proving how the topology of function spaces encodes the algorithmic and epistemic limits of prediction.
© 2025 Eduardo Dueñez, José Iovino, Tonatiuh Matos-Wiederhold, Luciano Salvetti, and Franklin D. Tall.
This repository contains the LaTeX source files for the paper Complexity of Deep Computations via Topology of Function Spaces, currently in preparation. It also contains the script tools/render_newton.py to make the diagrams, stored in images/, used in the paper.
These materials are shared for transparency and collaboration purposes only. Redistribution or reuse is not permitted without the authors' consent.