Preprint / scholarly article

Observation Quotients and Learning-as-Lifting

K. Takahashi

Published
DOI
10.5281/zenodo.17539800

Abstract

On probability measures, we study the inf-projection G of an energy F along an observation map O and prove an image-EVI result: lambda-EVI gradient flows of F on (P2(X), W2) push forward to relaxed or exact lambda-EVI flows of G on (P2(Z), W2).

Keywords

  • AI
  • large language models
  • geometry
  • information theory
  • optimal transport
  • machine learning
  • evi
  • jko

Identifiers and source records