Boosting as a Product of Experts

Narayanan Edakunni; Gavin Brown; TMD Kovacs

Boosting as a Product of Experts

Narayanan Edakunni
, Gavin Brown
, TMD Kovacs

Intelligent Systems Laboratory

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

4 Citations (Scopus)

Abstract

In this paper, we derive a novel probabilistic model of boosting as a Product of Experts. We re-derive the boosting algorithm as a greedy incremental model selection procedure which ensures that addition of new experts to the ensemble does not decrease the likelihood of the data. These learning rules lead to a generic boosting algorithm - POE-Boost which turns out to be similar to the AdaBoost algorithm under certain assumptions on the expert probabilities. The paper then extends the POEBoost algorithm to POEBoost.CS which handles hypothesis that produce probabilistic predictions. This new algorithm is shown to have better generalization performance compared to other state of the art algorithms.

Translated title of the contribution	Boosting as a Product of Experts
Original language	English
Title of host publication	Uncertainty in Artificial Intelligence
Publisher	Association for Uncertainty in Artificial Intelligence Press
Number of pages	8
Publication status	Published - 2011

Bibliographical note

Conference Proceedings/Title of Journal: Uncertainty in Artificial Intelligence. Proceedings of the Twenty-Seventh Conference
Conference Organiser: Peter Grünwald et al
Other: http://www.auai.org/

Access to Document

http://www.cs.bris.ac.uk/Publications/pub_master.jsp?id=2001427

Handle.net

Persistent link

Cite this

@inproceedings{0a8ee14db53b448f84449d2c512ef2ab,

title = "Boosting as a Product of Experts",

abstract = "In this paper, we derive a novel probabilistic model of boosting as a Product of Experts. We re-derive the boosting algorithm as a greedy incremental model selection procedure which ensures that addition of new experts to the ensemble does not decrease the likelihood of the data. These learning rules lead to a generic boosting algorithm - POE-Boost which turns out to be similar to the AdaBoost algorithm under certain assumptions on the expert probabilities. The paper then extends the POEBoost algorithm to POEBoost.CS which handles hypothesis that produce probabilistic predictions. This new algorithm is shown to have better generalization performance compared to other state of the art algorithms.",

author = "Narayanan Edakunni and Gavin Brown and TMD Kovacs",

note = "Conference Proceedings/Title of Journal: Uncertainty in Artificial Intelligence. Proceedings of the Twenty-Seventh Conference Conference Organiser: Peter Gr{\"u}nwald et al Other: http://www.auai.org/",

year = "2011",

language = "English",

booktitle = "Uncertainty in Artificial Intelligence",

publisher = "Association for Uncertainty in Artificial Intelligence Press",

}

TY - GEN

T1 - Boosting as a Product of Experts

AU - Edakunni, Narayanan

AU - Brown, Gavin

AU - Kovacs, TMD

N1 - Conference Proceedings/Title of Journal: Uncertainty in Artificial Intelligence. Proceedings of the Twenty-Seventh Conference Conference Organiser: Peter Grünwald et al Other: http://www.auai.org/

PY - 2011

Y1 - 2011

N2 - In this paper, we derive a novel probabilistic model of boosting as a Product of Experts. We re-derive the boosting algorithm as a greedy incremental model selection procedure which ensures that addition of new experts to the ensemble does not decrease the likelihood of the data. These learning rules lead to a generic boosting algorithm - POE-Boost which turns out to be similar to the AdaBoost algorithm under certain assumptions on the expert probabilities. The paper then extends the POEBoost algorithm to POEBoost.CS which handles hypothesis that produce probabilistic predictions. This new algorithm is shown to have better generalization performance compared to other state of the art algorithms.

AB - In this paper, we derive a novel probabilistic model of boosting as a Product of Experts. We re-derive the boosting algorithm as a greedy incremental model selection procedure which ensures that addition of new experts to the ensemble does not decrease the likelihood of the data. These learning rules lead to a generic boosting algorithm - POE-Boost which turns out to be similar to the AdaBoost algorithm under certain assumptions on the expert probabilities. The paper then extends the POEBoost algorithm to POEBoost.CS which handles hypothesis that produce probabilistic predictions. This new algorithm is shown to have better generalization performance compared to other state of the art algorithms.

M3 - Conference Contribution (Conference Proceeding)

BT - Uncertainty in Artificial Intelligence

PB - Association for Uncertainty in Artificial Intelligence Press

ER -

Boosting as a Product of Experts

Abstract

Bibliographical note

Access to Document

Handle.net

Fingerprint

Cite this