Estimating relatedness via data compression

Brendan Juba

Estimating relatedness via data compression

Brendan Juba

Department of Computer Science & Engineering

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

4 Scopus citations

Abstract

We show that it is possible to use data compression on independently obtained hypotheses from various tasks to algorithmically provide guarantees that the tasks are sufficiently related to benefit from multitask learning. We give uniform bounds in terms of the empirical average error for the true average error of the n hypotheses provided by deterministic learning algorithms drawing independent samples from a set of n unknown computable task distributions over finite sets.

Original language	English
Title of host publication	ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning
Pages	441-448
Number of pages	8
State	Published - 2006
Event	ICML 2006: 23rd International Conference on Machine Learning - Pittsburgh, PA, United States Duration: Jun 25 2006 → Jun 29 2006

Publication series

Name	ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning
Volume	2006

Conference

Conference	ICML 2006: 23rd International Conference on Machine Learning
Country/Territory	United States
City	Pittsburgh, PA
Period	06/25/06 → 06/29/06

Cite this

@inproceedings{d3585c90a41743ecb0faca07ef525b8c,

title = "Estimating relatedness via data compression",

abstract = "We show that it is possible to use data compression on independently obtained hypotheses from various tasks to algorithmically provide guarantees that the tasks are sufficiently related to benefit from multitask learning. We give uniform bounds in terms of the empirical average error for the true average error of the n hypotheses provided by deterministic learning algorithms drawing independent samples from a set of n unknown computable task distributions over finite sets.",

author = "Brendan Juba",

year = "2006",

language = "English",

isbn = "1595933832",

series = "ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning",

pages = "441--448",

booktitle = "ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning",

note = "ICML 2006: 23rd International Conference on Machine Learning ; Conference date: 25-06-2006 Through 29-06-2006",

}

TY - GEN

T1 - Estimating relatedness via data compression

AU - Juba, Brendan

PY - 2006

Y1 - 2006

N2 - We show that it is possible to use data compression on independently obtained hypotheses from various tasks to algorithmically provide guarantees that the tasks are sufficiently related to benefit from multitask learning. We give uniform bounds in terms of the empirical average error for the true average error of the n hypotheses provided by deterministic learning algorithms drawing independent samples from a set of n unknown computable task distributions over finite sets.

AB - We show that it is possible to use data compression on independently obtained hypotheses from various tasks to algorithmically provide guarantees that the tasks are sufficiently related to benefit from multitask learning. We give uniform bounds in terms of the empirical average error for the true average error of the n hypotheses provided by deterministic learning algorithms drawing independent samples from a set of n unknown computable task distributions over finite sets.

UR - https://www.scopus.com/pages/publications/33749369550

M3 - Conference contribution

AN - SCOPUS:33749369550

SN - 1595933832

SN - 9781595933836

T3 - ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning

SP - 441

EP - 448

BT - ICML 2006 - Proceedings of the 23rd International Conference on Machine Learning

T2 - ICML 2006: 23rd International Conference on Machine Learning

Y2 - 25 June 2006 through 29 June 2006

ER -

Estimating relatedness via data compression

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this