Learning linear classifiers sensitive to example dependent and noisy costs

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Authors

Learning algorithms from the fields of artificial neural networks and machine learning, typically, do not take any costs into account or allow only costs depending on the classes of the examples that are used for learning. As an extension of class dependent costs, we consider costs that are example, i.e. feature and class dependent. We derive a cost-sensitive perceptron learning rule for non-separable classes, that can be extended to multi-modal classes (DIPOL) and present a natural cost-sensitive extension of the support vector machine (SVM).
OriginalspracheEnglisch
ZeitschriftLecture Notes in Computer Science
Jahrgang2810
Seiten (von - bis)167-178
Anzahl der Seiten12
ISSN0302-9743
DOIs
PublikationsstatusErschienen - 01.01.2003
Extern publiziertJa

DOI

Zuletzt angesehen

Forschende

  1. Isabelle Fincke

Publikationen

  1. Not only biocidal products
  2. The role of the situation model in mathematical modelling
  3. Effects of grassland management, endophytic fungi and predators on aphid abundance in two distinct regions
  4. Microeconometric Studies on Firm Behavior and Performance
  5. Generalizing Trust
  6. Iconography on Scientific Instruments. Introduction
  7. Technology Implementation in Pre-Service Science Teacher Education Based on the Transformative View of TPACK: Effects on Pre-Service Teachers' TPACK, Behavioral Orientations and Actions in Practice
  8. Algorithmic Catastrophe - the Revenge of Contingency
  9. A duty-block network approach for an integrated driver rostering problem in public bus transport
  10. Towards a New Aesthetic
  11. The social dynamics of knowledge hiding
  12. Export entry, export exit, and productivity in German manufacturing industries
  13. A panel cointegrating rank test with structural breaks and cross-sectional dependence
  14. Creating a space for cooperation
  15. Active First Movers vs. Late Free-Riders? An Empirical Analysis of UN PRI Signatories' Commitment
  16. "It´s All in the Game"
  17. Usage pattern-based exposure screening as a simple tool for the regional priority-setting in environmental risk assessment of veterinary antibiotics
  18. Identifying core habitat before it's too late
  19. Creating Value from in-Vehicle Data
  20. Activity-based working
  21. Adjust for windows
  22. 'KNOW WHY' thinking as a new approach to systems thinking
  23. The Role of Network Size for the Robustness of Centrality Measures
  24. The case survey method and applications in political science
  25. Guest editorial
  26. Managing Global Production Networks
  27. Complexity Measures of Traffic Scenarios
  28. How problem-based or direct instructional case-based learning environments influence pre-service teachers’ cognitive load, motivation and emotions
  29. Does symbolic representation through class signalling appeal to voters? Evidence from a conjoint experiment
  30. Conditions of One-Way and Two-Way Approaches in Strategic Start-Up Communication
  31. Comparative study of resonant circuit for power transmission via inductive link
  32. Applying Necessity and Proportionality to Anti-Terrorist Self-Defence
  33. A Bayesian EAP-Based Nonlinear Extension of Croon and Van Veldhoven’s Model for Analyzing Data from Micro–Macro Multilevel Designs