On efficiency of Learning Under Privileged Information

Ilia Nouretdinov

On efficiency of Learning Under Privileged Information

Department of Computer Science

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

Abstract

The paradigm of Learning Under Privileged Information (LUPI) was used in various prac- tical applications, including its combination with Conformal Prediction (CP) framework. In this note, we discuss possible sources and limitations of its efficiency. We try to argue that accuracy improvement coming from using privileged information is not occasional. For this goal, we consider some minimalistic models of LUPI where the contribution of the privileged information appears in its noise-free essence. Then, we discuss connection of LUPI paradigm and CP framework in relation with the models.

Original language	English
Title of host publication	11th Symposium on Conformal and Probabilistic Prediction with Applications
Place of Publication	Brighton, UK
Publisher	Proceedings of Machine Learning Research, 2022 Conformal and Probabilistic Prediction and Applications
Number of pages	14
Volume	179
Publication status	Published - 25 Aug 2022

Keywords

Conformal prediction
privileged information
efficiency

Access to Document

https://copa-conference.com/papers/COPA2022_paper_14.pdfLicence: Other

Cite this

@inproceedings{e0a7e44b42b941f1bba315216feea024,

title = "On efficiency of Learning Under Privileged Information",

abstract = "The paradigm of Learning Under Privileged Information (LUPI) was used in various prac- tical applications, including its combination with Conformal Prediction (CP) framework. In this note, we discuss possible sources and limitations of its efficiency. We try to argue that accuracy improvement coming from using privileged information is not occasional. For this goal, we consider some minimalistic models of LUPI where the contribution of the privileged information appears in its noise-free essence. Then, we discuss connection of LUPI paradigm and CP framework in relation with the models.",

keywords = "Conformal prediction, privileged information, efficiency",

author = "Ilia Nouretdinov",

year = "2022",

month = aug,

day = "25",

language = "English",

volume = "179",

booktitle = "11th Symposium on Conformal and Probabilistic Prediction with Applications",

publisher = "Proceedings of Machine Learning Research, 2022 Conformal and Probabilistic Prediction and Applications",

}

On efficiency of Learning Under Privileged Information. / Nouretdinov, Ilia.
11th Symposium on Conformal and Probabilistic Prediction with Applications. Vol. 179 Brighton, UK: Proceedings of Machine Learning Research, 2022 Conformal and Probabilistic Prediction and Applications, 2022.

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

TY - GEN

T1 - On efficiency of Learning Under Privileged Information

AU - Nouretdinov, Ilia

PY - 2022/8/25

Y1 - 2022/8/25

N2 - The paradigm of Learning Under Privileged Information (LUPI) was used in various prac- tical applications, including its combination with Conformal Prediction (CP) framework. In this note, we discuss possible sources and limitations of its efficiency. We try to argue that accuracy improvement coming from using privileged information is not occasional. For this goal, we consider some minimalistic models of LUPI where the contribution of the privileged information appears in its noise-free essence. Then, we discuss connection of LUPI paradigm and CP framework in relation with the models.

AB - The paradigm of Learning Under Privileged Information (LUPI) was used in various prac- tical applications, including its combination with Conformal Prediction (CP) framework. In this note, we discuss possible sources and limitations of its efficiency. We try to argue that accuracy improvement coming from using privileged information is not occasional. For this goal, we consider some minimalistic models of LUPI where the contribution of the privileged information appears in its noise-free essence. Then, we discuss connection of LUPI paradigm and CP framework in relation with the models.

KW - Conformal prediction

KW - privileged information

KW - efficiency

M3 - Conference contribution

VL - 179

BT - 11th Symposium on Conformal and Probabilistic Prediction with Applications

PB - Proceedings of Machine Learning Research, 2022 Conformal and Probabilistic Prediction and Applications

CY - Brighton, UK

ER -