
-
Quantitative Rates and Fundamental Obstructions to Non-Euclidean Universal Approximation with Deep Narrow Feed-Forward Networks
By incorporating structured pairs of non-trainable input and output laye...
read it
-
Optimizing Optimizers: Regret-optimal gradient descent algorithms
The need for fast and robust optimization algorithms are of critical imp...
read it
-
Overcoming The Limitations of Neural Networks in Composite-Pattern Learning with Architopes
The effectiveness of neural networks in solving complex problems is well...
read it
-
Architopes: An Architecture Modification for Composite Pattern Learning, Increased Expressiveness, and Reduced Training Time
We introduce a simple neural network architecture modification that enab...
read it
-
Non-Euclidean Universal Approximation
Modifications to a neural network's input and output layers are often re...
read it
-
Universal Approximation Theorems
The universal approximation theorem established the density of specific ...
read it
-
The NEU Meta-Algorithm for Geometric Learning with Applications in Finance
We introduce a meta-algorithm, called non-Euclidean upgrading (NEU), whi...
read it
-
Geometric Learning and Filtering in Finance
We develop a method for incorporating relevant non-Euclidean geometric i...
read it
-
Arbitrage-Free Regularization
We introduce a path-dependent geometric framework which generalizes the ...
read it