Batch query strategies for one-class active learning: Unterschied zwischen den Versionen

Aus IPD-Institutsseminar
Zur Navigation springen Zur Suche springen
 
Zeile 5: Zeile 5:
 
|betreuer=Holger Trittenbach
 
|betreuer=Holger Trittenbach
 
|termin=Institutsseminar/2018-10-26
 
|termin=Institutsseminar/2018-10-26
|kurzfassung=One-class classifiers learn to distinguish normal objects from outliers. Such classifiers are therefore suitable for strongly imbalanced class distributions with only a small fraction of outliers. There are extensions of one-class classifiers that make use of labeled samples to improve classification quality. As the labeling must often be done by an expert, obtaining these labeled samples can be very expensive and time-consuming. To reduce these labeling efforts, one may rely on active learning methods that detect samples where obtaining a label from the user is worthwhile, with the goal of reducing the labeling effort to a fraction of the original data set.
+
|kurzfassung=One-class classifiers learn to distinguish normal objects from outliers. These classifiers are therefore suitable for strongly imbalanced class distributions with only a small fraction of outliers. Extensions of one-class classifiers make use of labeled samples to improve classification quality. As this labeling process is often time-consuming, one may use active learning methods to detect samples where obtaining a label from the user is worthwhile, with the goal of reducing the labeling effort to a fraction of the original data set. In the case of one-class active learning this labeling process consists of sequential queries, where the user labels one sample at a time. While batch queries where the user labels multiple samples at a time have potential advantages, for example parallelizing the labeling process, their application has so far been limited to binary and multi-class classification. In this thesis we explore whether batch queries can be used for one-class classification. We strive towards a novel batch query strategy for one-class classification by applying concepts from multi-class classification to the requirements of one-class active learning.
 
 
This labeling process can be performed with sequential queries, where the user is asked to label one sample at a time, or batch queries, where the user is asked to label multiple samples at once.
 
Research on Active Learning for one-class classifiers has focused on sequential queries. While batch queries have potential advantages, like enabling a parallelization of the labeling process, their application has so far been limited to binary and multi-class classification.
 
 
 
We propose a new batch query strategy for one-class classification by applying concepts from multi-class classification to the special requirements of one-class active learning.
 
 
}}
 
}}

Aktuelle Version vom 22. Oktober 2018, 11:06 Uhr

Vortragende(r) Dennis Vetter
Vortragstyp Proposal
Betreuer(in) Holger Trittenbach
Termin Fr 26. Oktober 2018
Kurzfassung One-class classifiers learn to distinguish normal objects from outliers. These classifiers are therefore suitable for strongly imbalanced class distributions with only a small fraction of outliers. Extensions of one-class classifiers make use of labeled samples to improve classification quality. As this labeling process is often time-consuming, one may use active learning methods to detect samples where obtaining a label from the user is worthwhile, with the goal of reducing the labeling effort to a fraction of the original data set. In the case of one-class active learning this labeling process consists of sequential queries, where the user labels one sample at a time. While batch queries where the user labels multiple samples at a time have potential advantages, for example parallelizing the labeling process, their application has so far been limited to binary and multi-class classification. In this thesis we explore whether batch queries can be used for one-class classification. We strive towards a novel batch query strategy for one-class classification by applying concepts from multi-class classification to the requirements of one-class active learning.