On the α-loss Landscape in the Logistic Model

Tyler Sypherd; Mario Diaz; Lalitha Sankar; Gautam Dasarathy

doi:10.1109/ISIT44484.2020.9174356

On the α-loss Landscape in the Logistic Model

Tyler Sypherd, Mario Diaz, Lalitha Sankar, Gautam Dasarathy

Electrical, Computer, and Energy Engineering, School of (IAFSE-ECEE)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

We analyze the optimization landscape of a recently introduced tunable class of loss functions called α-loss, α (0, ∞], in the logistic model. This family encapsulates the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞) and contains compelling properties that enable the practitioner to discern among a host of operating conditions relevant to emerging learning methods. Specifically, we study the evolution of the optimization landscape of α-loss with respect to α using tools drawn from the study of strictly-locally-quasi-convex functions in addition to geometric techniques. We interpret these results in terms of optimization complexity via normalized gradient descent.

Original language	English (US)
Title of host publication	2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2700-2705
Number of pages	6
ISBN (Electronic)	9781728164328
DOIs	https://doi.org/10.1109/ISIT44484.2020.9174356
State	Published - Jun 2020
Event	2020 IEEE International Symposium on Information Theory, ISIT 2020 - Los Angeles, United States Duration: Jul 21 2020 → Jul 26 2020

Publication series

Name	IEEE International Symposium on Information Theory - Proceedings
Volume	2020-June
ISSN (Print)	2157-8095

Conference

Conference	2020 IEEE International Symposium on Information Theory, ISIT 2020
Country/Territory	United States
City	Los Angeles
Period	7/21/20 → 7/26/20

ASJC Scopus subject areas

Theoretical Computer Science
Information Systems
Modeling and Simulation
Applied Mathematics

Access to Document

10.1109/ISIT44484.2020.9174356

Cite this

Sypherd, T., Diaz, M., Sankar, L., & Dasarathy, G. (2020). On the α-loss Landscape in the Logistic Model. In 2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings (pp. 2700-2705). Article 9174356 (IEEE International Symposium on Information Theory - Proceedings; Vol. 2020-June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISIT44484.2020.9174356

On the α-loss Landscape in the Logistic Model. / Sypherd, Tyler; Diaz, Mario; Sankar, Lalitha et al.
2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. p. 2700-2705 9174356 (IEEE International Symposium on Information Theory - Proceedings; Vol. 2020-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Sypherd, T, Diaz, M, Sankar, L & Dasarathy, G 2020, On the α-loss Landscape in the Logistic Model. in 2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings., 9174356, IEEE International Symposium on Information Theory - Proceedings, vol. 2020-June, Institute of Electrical and Electronics Engineers Inc., pp. 2700-2705, 2020 IEEE International Symposium on Information Theory, ISIT 2020, Los Angeles, United States, 7/21/20. https://doi.org/10.1109/ISIT44484.2020.9174356

@inproceedings{ef0d3abd457e4c64907d7dabe782cc94,

title = "On the α-loss Landscape in the Logistic Model",

abstract = "We analyze the optimization landscape of a recently introduced tunable class of loss functions called α-loss, α (0, ∞], in the logistic model. This family encapsulates the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞) and contains compelling properties that enable the practitioner to discern among a host of operating conditions relevant to emerging learning methods. Specifically, we study the evolution of the optimization landscape of α-loss with respect to α using tools drawn from the study of strictly-locally-quasi-convex functions in addition to geometric techniques. We interpret these results in terms of optimization complexity via normalized gradient descent.",

author = "Tyler Sypherd and Mario Diaz and Lalitha Sankar and Gautam Dasarathy",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 2020 IEEE International Symposium on Information Theory, ISIT 2020 ; Conference date: 21-07-2020 Through 26-07-2020",

year = "2020",

month = jun,

doi = "10.1109/ISIT44484.2020.9174356",

language = "English (US)",

series = "IEEE International Symposium on Information Theory - Proceedings",

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

pages = "2700--2705",

booktitle = "2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings",

}

TY - GEN

T1 - On the α-loss Landscape in the Logistic Model

AU - Sypherd, Tyler

AU - Diaz, Mario

AU - Sankar, Lalitha

AU - Dasarathy, Gautam

PY - 2020/6

Y1 - 2020/6

N2 - We analyze the optimization landscape of a recently introduced tunable class of loss functions called α-loss, α (0, ∞], in the logistic model. This family encapsulates the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞) and contains compelling properties that enable the practitioner to discern among a host of operating conditions relevant to emerging learning methods. Specifically, we study the evolution of the optimization landscape of α-loss with respect to α using tools drawn from the study of strictly-locally-quasi-convex functions in addition to geometric techniques. We interpret these results in terms of optimization complexity via normalized gradient descent.

AB - We analyze the optimization landscape of a recently introduced tunable class of loss functions called α-loss, α (0, ∞], in the logistic model. This family encapsulates the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞) and contains compelling properties that enable the practitioner to discern among a host of operating conditions relevant to emerging learning methods. Specifically, we study the evolution of the optimization landscape of α-loss with respect to α using tools drawn from the study of strictly-locally-quasi-convex functions in addition to geometric techniques. We interpret these results in terms of optimization complexity via normalized gradient descent.

UR - http://www.scopus.com/inward/record.url?scp=85090404551&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090404551&partnerID=8YFLogxK

U2 - 10.1109/ISIT44484.2020.9174356

DO - 10.1109/ISIT44484.2020.9174356

M3 - Conference contribution

AN - SCOPUS:85090404551

T3 - IEEE International Symposium on Information Theory - Proceedings

SP - 2700

EP - 2705

BT - 2020 IEEE International Symposium on Information Theory, ISIT 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE International Symposium on Information Theory, ISIT 2020

Y2 - 21 July 2020 through 26 July 2020

ER -

On the α-loss Landscape in the Logistic Model

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this