Development of an Active Learning Approach for One Class Classifi cation using Bayesian Uncertainty: Unterschied zwischen den Versionen

Aus SDQ-Institutsseminar
(Die Seite wurde neu angelegt: „{{Vortrag |vortragender=Tobias Haßberg |email=uferh@student.kit.edu |vortragstyp=Masterarbeit |betreuer=Bela Böhnke |termin=Institutsseminar/2022-06-03 |vort…“)
 
Keine Bearbeitungszusammenfassung
 
Zeile 4: Zeile 4:
|vortragstyp=Masterarbeit
|vortragstyp=Masterarbeit
|betreuer=Bela Böhnke
|betreuer=Bela Böhnke
|termin=Institutsseminar/2022-06-03
|termin=Institutsseminar/2022-06-03 Zusatztermin
|vortragsmodus=in Präsenz
|vortragsmodus=in Präsenz
|kurzfassung=In One-Class classification, the classifier decides if points belong to a specific class. In this thesis, we propose an One-Class classification approach, suitable for active learning, that models for each point, a prediction range in which the model assumes the points state to be. The proposed classifier uses a Gaussian process. We use the Gaussian processes prediction range to derive a certainty measure, that considers the available labeled points for stating its certainty. We compared this approach against baseline classifiers and show the correlation between the classifier's uncertainty and misclassification ratio.
|kurzfassung=In One-Class classification, the classifier decides if points belong to a specific class. In this thesis, we propose an One-Class classification approach, suitable for active learning, that models for each point, a prediction range in which the model assumes the points state to be. The proposed classifier uses a Gaussian process. We use the Gaussian processes prediction range to derive a certainty measure, that considers the available labeled points for stating its certainty. We compared this approach against baseline classifiers and show the correlation between the classifier's uncertainty and misclassification ratio.
}}
}}

Aktuelle Version vom 24. Mai 2022, 10:37 Uhr

Vortragende(r) Tobias Haßberg
Vortragstyp Masterarbeit
Betreuer(in) Bela Böhnke
Termin Fr 3. Juni 2022
Vortragsmodus in Präsenz
Kurzfassung In One-Class classification, the classifier decides if points belong to a specific class. In this thesis, we propose an One-Class classification approach, suitable for active learning, that models for each point, a prediction range in which the model assumes the points state to be. The proposed classifier uses a Gaussian process. We use the Gaussian processes prediction range to derive a certainty measure, that considers the available labeled points for stating its certainty. We compared this approach against baseline classifiers and show the correlation between the classifier's uncertainty and misclassification ratio.