Introduces models for neural ODEs trained with online SGD and derives their high-dimensional learning curves via dynamical mean field theory.
A mean-field optimal control formulation of deep learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Deep networks are framed as memory spaces whose complexity is defined by a Fisher metric, with the least action principle linking this complexity to generalization and disentanglement for better interpretability.
citing papers explorer
-
Theory of learning of high-dimensional controlled non-linear dynamical systems (I): models and methods
Introduces models for neural ODEs trained with online SGD and derives their high-dimensional learning curves via dynamical mean field theory.
-
Deep network as memory space: complexity, generalization, disentangled representation and interpretability
Deep networks are framed as memory spaces whose complexity is defined by a Fisher metric, with the least action principle linking this complexity to generalization and disentanglement for better interpretability.