Beyond novelty detection: Incongruent events, when general and specific classifiers disagree

Daphna Weinshall, Hynek Hermansky, Alon Zweig, Jie Luo, Holly Jimison, Frank Ohl, Misha Pavel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Citations (Scopus)

Abstract

Unexpected stimuli are a challenge to any machine learning algorithm. Here we identify distinct types of unexpected events, focusing on 'incongruent events' - when 'general level' and 'specific level' classifiers give conflicting predictions. We define a formal framework for the representation and processing of incongruent events: starting from the notion of label hierarchy, we show how partial order on labels can be deduced from such hierarchies. For each event, we compute its probability in different ways, based on adjacent levels (according to the partial order) in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level (in accordance with the partial order) is much smaller than the probability computed based on some more general level, leading to conflicting predictions. We derive algorithms to detect incongruent events from different types of hierarchies, corresponding to class membership or part membership. Respectively, we show promising results with real data on two specific problems: Out Of Vocabulary words in speech recognition, and the identification of a new sub-class (e.g., the face of a new individual) in audio-visual facial object recognition.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference
Pages1745-1752
Number of pages8
StatePublished - 2009
Event22nd Annual Conference on Neural Information Processing Systems, NIPS 2008 - Vancouver, BC, Canada
Duration: Dec 8 2008Dec 11 2008

Other

Other22nd Annual Conference on Neural Information Processing Systems, NIPS 2008
CountryCanada
CityVancouver, BC
Period12/8/0812/11/08

Fingerprint

Labels
Classifiers
Object recognition
Speech recognition
Learning algorithms
Learning systems
Processing

ASJC Scopus subject areas

  • Information Systems

Cite this

Weinshall, D., Hermansky, H., Zweig, A., Luo, J., Jimison, H., Ohl, F., & Pavel, M. (2009). Beyond novelty detection: Incongruent events, when general and specific classifiers disagree. In Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference (pp. 1745-1752)

Beyond novelty detection : Incongruent events, when general and specific classifiers disagree. / Weinshall, Daphna; Hermansky, Hynek; Zweig, Alon; Luo, Jie; Jimison, Holly; Ohl, Frank; Pavel, Misha.

Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. p. 1745-1752.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Weinshall, D, Hermansky, H, Zweig, A, Luo, J, Jimison, H, Ohl, F & Pavel, M 2009, Beyond novelty detection: Incongruent events, when general and specific classifiers disagree. in Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. pp. 1745-1752, 22nd Annual Conference on Neural Information Processing Systems, NIPS 2008, Vancouver, BC, Canada, 12/8/08.
Weinshall D, Hermansky H, Zweig A, Luo J, Jimison H, Ohl F et al. Beyond novelty detection: Incongruent events, when general and specific classifiers disagree. In Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. p. 1745-1752
Weinshall, Daphna ; Hermansky, Hynek ; Zweig, Alon ; Luo, Jie ; Jimison, Holly ; Ohl, Frank ; Pavel, Misha. / Beyond novelty detection : Incongruent events, when general and specific classifiers disagree. Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. pp. 1745-1752
@inproceedings{0f26f20441c04502b775d200a89d0ffc,
title = "Beyond novelty detection: Incongruent events, when general and specific classifiers disagree",
abstract = "Unexpected stimuli are a challenge to any machine learning algorithm. Here we identify distinct types of unexpected events, focusing on 'incongruent events' - when 'general level' and 'specific level' classifiers give conflicting predictions. We define a formal framework for the representation and processing of incongruent events: starting from the notion of label hierarchy, we show how partial order on labels can be deduced from such hierarchies. For each event, we compute its probability in different ways, based on adjacent levels (according to the partial order) in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level (in accordance with the partial order) is much smaller than the probability computed based on some more general level, leading to conflicting predictions. We derive algorithms to detect incongruent events from different types of hierarchies, corresponding to class membership or part membership. Respectively, we show promising results with real data on two specific problems: Out Of Vocabulary words in speech recognition, and the identification of a new sub-class (e.g., the face of a new individual) in audio-visual facial object recognition.",
author = "Daphna Weinshall and Hynek Hermansky and Alon Zweig and Jie Luo and Holly Jimison and Frank Ohl and Misha Pavel",
year = "2009",
language = "English (US)",
isbn = "9781605609492",
pages = "1745--1752",
booktitle = "Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference",

}

TY - GEN

T1 - Beyond novelty detection

T2 - Incongruent events, when general and specific classifiers disagree

AU - Weinshall, Daphna

AU - Hermansky, Hynek

AU - Zweig, Alon

AU - Luo, Jie

AU - Jimison, Holly

AU - Ohl, Frank

AU - Pavel, Misha

PY - 2009

Y1 - 2009

N2 - Unexpected stimuli are a challenge to any machine learning algorithm. Here we identify distinct types of unexpected events, focusing on 'incongruent events' - when 'general level' and 'specific level' classifiers give conflicting predictions. We define a formal framework for the representation and processing of incongruent events: starting from the notion of label hierarchy, we show how partial order on labels can be deduced from such hierarchies. For each event, we compute its probability in different ways, based on adjacent levels (according to the partial order) in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level (in accordance with the partial order) is much smaller than the probability computed based on some more general level, leading to conflicting predictions. We derive algorithms to detect incongruent events from different types of hierarchies, corresponding to class membership or part membership. Respectively, we show promising results with real data on two specific problems: Out Of Vocabulary words in speech recognition, and the identification of a new sub-class (e.g., the face of a new individual) in audio-visual facial object recognition.

AB - Unexpected stimuli are a challenge to any machine learning algorithm. Here we identify distinct types of unexpected events, focusing on 'incongruent events' - when 'general level' and 'specific level' classifiers give conflicting predictions. We define a formal framework for the representation and processing of incongruent events: starting from the notion of label hierarchy, we show how partial order on labels can be deduced from such hierarchies. For each event, we compute its probability in different ways, based on adjacent levels (according to the partial order) in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level (in accordance with the partial order) is much smaller than the probability computed based on some more general level, leading to conflicting predictions. We derive algorithms to detect incongruent events from different types of hierarchies, corresponding to class membership or part membership. Respectively, we show promising results with real data on two specific problems: Out Of Vocabulary words in speech recognition, and the identification of a new sub-class (e.g., the face of a new individual) in audio-visual facial object recognition.

UR - http://www.scopus.com/inward/record.url?scp=84855184457&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84855184457&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84855184457

SN - 9781605609492

SP - 1745

EP - 1752

BT - Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference

ER -