An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies

Yury Kalnishkan

doi:10.1007/978-3-319-46379-7_16

An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies

Yury Kalnishkan

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

336 Downloads (Pure)

Abstract

The paper presents a competitive prediction-style upper bound on the square loss of the Aggregating Algorithm for Regression with Changing Dependencies in the linear case. The algorithm is able to compete with a sequence of linear predictors provided the sum of squared Euclidean norms of differences of regression coefficient vectors grows at a sublinear rate.

Original language	English
Title of host publication	Algorithmic Learning Theory
Subtitle of host publication	27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings
Publisher	Springer International Publishing
Pages	238-252
Number of pages	15
Volume	9925
ISBN (Electronic)	978-3-319-46379-7
ISBN (Print)	978-3-319-46378-0
DOIs	https://doi.org/10.1007/978-3-319-46379-7_16
Publication status	Published - 2016

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer International Publishing
Volume	9925
ISSN (Print)	0302-9743

Keywords

on-line learning
prediction with expert advice
competitive prediction
linear regression

Access to Document

10.1007/978-3-319-46379-7_16

Accepted ManuscriptAccepted author manuscript, 399 KB

http://link.springer.com/chapter/10.1007%2F978-3-319-46379-7_16

Cite this

Kalnishkan, Y. (2016). An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies. In Algorithmic Learning Theory : 27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings (Vol. 9925, pp. 238-252). (Lecture Notes in Computer Science; Vol. 9925). Springer International Publishing. https://doi.org/10.1007/978-3-319-46379-7_16

@inproceedings{dc9dd9d8a831453aa5adafbf5082c9c3,

title = "An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies",

abstract = "The paper presents a competitive prediction-style upper bound on the square loss of the Aggregating Algorithm for Regression with Changing Dependencies in the linear case. The algorithm is able to compete with a sequence of linear predictors provided the sum of squared Euclidean norms of differences of regression coefficient vectors grows at a sublinear rate.",

keywords = "on-line learning, prediction with expert advice, competitive prediction, linear regression",

author = "Yury Kalnishkan",

year = "2016",

doi = "10.1007/978-3-319-46379-7_16",

language = "English",

isbn = "978-3-319-46378-0",

volume = "9925",

series = "Lecture Notes in Computer Science",

publisher = "Springer International Publishing",

pages = "238--252",

booktitle = "Algorithmic Learning Theory",

}

Kalnishkan, Y 2016, An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies. in Algorithmic Learning Theory : 27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings. vol. 9925, Lecture Notes in Computer Science, vol. 9925, Springer International Publishing, pp. 238-252. https://doi.org/10.1007/978-3-319-46379-7_16

An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies. / Kalnishkan, Yury.
Algorithmic Learning Theory : 27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings. Vol. 9925 Springer International Publishing, 2016. p. 238-252 (Lecture Notes in Computer Science; Vol. 9925).

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

TY - GEN

T1 - An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies

AU - Kalnishkan, Yury

PY - 2016

Y1 - 2016

N2 - The paper presents a competitive prediction-style upper bound on the square loss of the Aggregating Algorithm for Regression with Changing Dependencies in the linear case. The algorithm is able to compete with a sequence of linear predictors provided the sum of squared Euclidean norms of differences of regression coefficient vectors grows at a sublinear rate.

AB - The paper presents a competitive prediction-style upper bound on the square loss of the Aggregating Algorithm for Regression with Changing Dependencies in the linear case. The algorithm is able to compete with a sequence of linear predictors provided the sum of squared Euclidean norms of differences of regression coefficient vectors grows at a sublinear rate.

KW - on-line learning

KW - prediction with expert advice

KW - competitive prediction

KW - linear regression

UR - http://www.comp.nus.edu.sg/~fstephan/alt/alt2016/index.html

U2 - 10.1007/978-3-319-46379-7_16

DO - 10.1007/978-3-319-46379-7_16

M3 - Conference contribution

SN - 978-3-319-46378-0

VL - 9925

T3 - Lecture Notes in Computer Science

SP - 238

EP - 252

BT - Algorithmic Learning Theory

PB - Springer International Publishing

ER -

Kalnishkan Y. An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies. In Algorithmic Learning Theory : 27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings. Vol. 9925. Springer International Publishing. 2016. p. 238-252. (Lecture Notes in Computer Science). Epub 2016 Sept 21. doi: 10.1007/978-3-319-46379-7_16

An Upper Bound for Aggregating Algorithm for Regression with Changing Dependencies

Abstract

Publication series

Keywords

Access to Document

Other files and links

Cite this