On Local Entropy, Stochastic Control, and Deep Neural Networks

Michele Pavon

doi:10.1109/LCSYS.2022.3189927

On Local Entropy, Stochastic Control, and Deep Neural Networks

Michele Pavon

Mathematics

Research output: Contribution to journal › Article › peer-review

Abstract

In this letter, we connect some recent papers on smoothing of energy landscapes and scored-based generative models of machine learning to classical work in stochastic control. We clarify these connections providing rigorous statements and representations which may serve as guidelines for further learning models.

Original language	English (US)
Pages (from-to)	437-441
Number of pages	5
Journal	IEEE Control Systems Letters
Volume	7
DOIs	https://doi.org/10.1109/LCSYS.2022.3189927
State	Published - 2023

Keywords

Stochastic optimal control
machine learning
neural networks

ASJC Scopus subject areas

Control and Systems Engineering
Control and Optimization

Access to Document

10.1109/LCSYS.2022.3189927

Cite this

@article{61c67982c7e54d608c13e8d0bc427707,

title = "On Local Entropy, Stochastic Control, and Deep Neural Networks",

abstract = "In this letter, we connect some recent papers on smoothing of energy landscapes and scored-based generative models of machine learning to classical work in stochastic control. We clarify these connections providing rigorous statements and representations which may serve as guidelines for further learning models.",

keywords = "Stochastic optimal control, machine learning, neural networks",

author = "Michele Pavon",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.",

year = "2023",

doi = "10.1109/LCSYS.2022.3189927",

language = "English (US)",

volume = "7",

pages = "437--441",

journal = "IEEE Control Systems Letters",

issn = "2475-1456",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - On Local Entropy, Stochastic Control, and Deep Neural Networks

AU - Pavon, Michele

PY - 2023

Y1 - 2023

N2 - In this letter, we connect some recent papers on smoothing of energy landscapes and scored-based generative models of machine learning to classical work in stochastic control. We clarify these connections providing rigorous statements and representations which may serve as guidelines for further learning models.

AB - In this letter, we connect some recent papers on smoothing of energy landscapes and scored-based generative models of machine learning to classical work in stochastic control. We clarify these connections providing rigorous statements and representations which may serve as guidelines for further learning models.

KW - Stochastic optimal control

KW - machine learning

KW - neural networks

UR - http://www.scopus.com/inward/record.url?scp=85134261387&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85134261387&partnerID=8YFLogxK

U2 - 10.1109/LCSYS.2022.3189927

DO - 10.1109/LCSYS.2022.3189927

M3 - Article

AN - SCOPUS:85134261387

SN - 2475-1456

VL - 7

SP - 437

EP - 441

JO - IEEE Control Systems Letters

JF - IEEE Control Systems Letters

ER -

On Local Entropy, Stochastic Control, and Deep Neural Networks

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this