Deep Conditional Measure Quantization - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Deep Conditional Measure Quantization

Résumé

Quantization of a probability measure means representing it with a finite set of Dirac masses that approximates the input distribution well enough (in some metric space of probability measures). Various methods exists to do so, but the situation of quantizing a conditional law has been less explored. We propose a method, called DCMQ, involving a Huber-energy kernel-based approach coupled with a deep neural network architecture. The method is tested on several examples and obtains promising results.

Dates et versions

hal-04448823 , version 1 (09-02-2024)

Identifiants

Citer

Gabriel Turinici. Deep Conditional Measure Quantization. Optimization, Learning Algorithms and Applications, Sep 2023, Ponta Delgada, Portugal. pp.343-354, ⟨10.1007/978-3-031-53036-4_24⟩. ⟨hal-04448823⟩
6 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More