TY - JOUR
T1 - Beyond novelty detection
T2 - Incongruent events, when general and specific classifiers disagree
AU - Weinshall, Daphna
AU - Zweig, Alon
AU - Hermansky, Hynek
AU - Kombrink, Stefan
AU - Ohl, Frank W.
AU - Anemüller, Jörn
AU - Bach, Jörg Hendrik
AU - Van Gool, Luc
AU - Nater, Fabian
AU - Pajdla, Tomas
AU - Havlena, Michal
AU - Pavel, Misha
N1 - Funding Information:
This study was supported by the European Union under the DIRAC integrated project IST-027787.
PY - 2012
Y1 - 2012
N2 - Unexpected stimuli are a challenge to any machine learning algorithm. Here, we identify distinct types of unexpected 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 in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level is much smaller than the probability computed based on some more general level, leading to conflicting predictions. Algorithms are derived to detect incongruent events from different types of hierarchies, different applications, and a variety of data types. We present promising results for the detection of novel visual and audio objects, and new patterns of motion in video. We also discuss the detection of Out-Of-Vocabulary words in speech recognition, and the detection of incongruent events in a multimodal audiovisual scenario.
AB - Unexpected stimuli are a challenge to any machine learning algorithm. Here, we identify distinct types of unexpected 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 in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level is much smaller than the probability computed based on some more general level, leading to conflicting predictions. Algorithms are derived to detect incongruent events from different types of hierarchies, different applications, and a variety of data types. We present promising results for the detection of novel visual and audio objects, and new patterns of motion in video. We also discuss the detection of Out-Of-Vocabulary words in speech recognition, and the detection of incongruent events in a multimodal audiovisual scenario.
KW - Novelty detection
KW - categorization
KW - object recognition
KW - out-of-vocabulary words
UR - http://www.scopus.com/inward/record.url?scp=84865320756&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2011.279
DO - 10.1109/TPAMI.2011.279
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 22213766
AN - SCOPUS:84865320756
SN - 0162-8828
VL - 34
SP - 1886
EP - 1901
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 10
M1 - 6122029
ER -