Active and semi-supervised data domain description

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Authors

Data domain description techniques aim at deriving concise descriptions of objects belonging to a category of interest. For instance, the support vector domain description (SVDD) learns a hypersphere enclosing the bulk of provided unlabeled data such that points lying outside of the ball are considered anomalous. However, relevant information such as expert and background knowledge remain unused in the unsupervised setting. In this paper, we rephrase data domain description as a semi-supervised learning task, that is, we propose a semi-supervised generalization of data domain description (SSSVDD) to process unlabeled and labeled examples. The corresponding optimization problem is non-convex. We translate it into an unconstraint, continuous problem that can be optimized accurately by gradient-based techniques. Furthermore, we devise an effective active learning strategy to query low-confidence observations. Our empirical evaluation on network intrusion detection and object recognition tasks shows that our SSSVDDs consistently outperform baseline methods in relevant learning settings.
OriginalspracheEnglisch
TitelMachine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I
HerausgeberWray Buntine, Marko Grobelnik, Dunja Mladenic, John Shawe-Taylor
Anzahl der Seiten16
ErscheinungsortBerlin, Heidelberg
VerlagSpringer Verlag
Erscheinungsdatum01.07.2009
Seiten407-422
ISBN (Print)978-3-642-04179-2
ISBN (elektronisch)978-3-642-04180-8
DOIs
PublikationsstatusErschienen - 01.07.2009
Extern publiziertJa
VeranstaltungEuropean Conference on Machine Learning and Knowledge Discovery in Databases - 2009 - Bled, Slowenien
Dauer: 07.09.200911.09.2009
https://www.k4all.org/event/european-conference-on-machine-learning-and-principles-and-practice-of-knowledge-discovery-in-databases/

DOI

Zuletzt angesehen

Publikationen

  1. Double-fading support - A training approach to complex software systems
  2. Trajectory tracking using MPC and a velocity observer for flat actuator systems in automotive applications
  3. ℓp-norm multiple kernel learning
  4. Individual Scans Fusion in Virtual Knowledge Base for Navigation of Mobile Robotic Group with 3D TVS
  5. Improve a 3D distance measurement accuracy in stereo vision systems using optimization methods’ approach
  6. Functional Richness and Relative Resilience of Bird Communities in Regions with Different Land Use Intensities
  7. Pressure fault recognition and compensation with an adaptive feedforward regulator in a controlled hybrid actuator within engine applications
  8. The impact of goal focus, task type and group size on synchronous net-based collaborative learning discourses
  9. Material flow analysis between dynamic modelling and life cycle assessment
  10. Active plasma resonance spectroscopy: Eigenfunction solutions in spherical geometry
  11. Collaborative benchmarking of functional-structural root architecture models
  12. Wavelet functions for rejecting spurious values
  13. Finite element modeling of laser beam welding for residual stress calculation
  14. Internet research differs from research on internet users
  15. Frame-based Data Factorizations
  16. Strengthening the transformative impulse while mainstreaming real-world labs: Lessons learned from three years of BaWü-Labs
  17. Gerbil – Benchmarking named entity recognition and linking consistently
  18. Introduction: Habitual Action, Automaticity, and Control
  19. Practice and carryover effects when using small interaction devices
  20. Teaching Sustainable Development in a Sensory and Artful Way — Concepts, Methods, and Examples
  21. Influence of Mg content in Al alloys on processing characteristics and dynamically recrystallized microstructure of friction surfacing deposits
  22. Stimulating Computing
  23. Comparison of three methods of length compensation in a parallel kinematic and their equivalence conditions
  24. Can a Revision of the Universal Service Scope Result in Substantive Change?
  25. Modeling and simulation of the heterogenous material behavior in thermal-sprayed coatings
  26. Sliding Mode Control of an Inductive Power Transmission System with Maximum Efficiency
  27. Short-arc measurement and fitting based on the bidirectional prediction of observed data
  28. Graph-Based Early-Fusion for Flood Detection