Header logo is

Machine Learning Applied to Perception: Decision Images for Classification

2005

Conference Paper

ei


We study gender discrimination of human faces using a combination of psychophysical classification and discrimination experiments together with methods from machine learning. We reduce the dimensionality of a set of face images using principal component analysis, and then train a set of linear classifiers on this reduced representation (linear support vector machines (SVMs), relevance vector machines (RVMs), Fisher linear discriminant (FLD), and prototype (prot) classifiers) using human classification data. Because we combine a linear preprocessor with linear classifiers, the entire system acts as a linear classifier, allowing us to visualise the decision-image corresponding to the normal vector of the separating hyperplanes (SH) of each classifier. We predict that the female-to-maleness transition along the normal vector for classifiers closely mimicking human classification (SVM and RVM 1) should be faster than the transition along any other direction. A psychophysical discrimination experiment using the decision images as stimuli is consistent with this prediction.

Author(s): Wichmann, FA. and Graf, ABA. and Simoncelli, EP. and Bülthoff, HH. and Schölkopf, B.
Book Title: Advances in Neural Information Processing Systems 17
Journal: Advances in Neural Information Processing Systems
Pages: 1489-1496
Year: 2005
Month: July
Day: 0
Editors: LK, Saul and Y, Weiss and L, Bottou
Publisher: MIT Press

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: 18th Annual Conference on Neural Information Processing Systems (NIPS 2004)
Event Place: Vancouver, BC, Canada

Address: Cambridge, MA, USA
Digital: 0
ISBN: 0-262-19534-8
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{2784,
  title = {Machine Learning Applied to Perception: Decision Images for Classification},
  author = {Wichmann, FA. and Graf, ABA. and Simoncelli, EP. and B{\"u}lthoff, HH. and Sch{\"o}lkopf, B.},
  journal = {Advances in Neural Information Processing Systems},
  booktitle = {Advances in Neural Information Processing Systems 17},
  pages = {1489-1496},
  editors = {LK, Saul and Y, Weiss and L, Bottou},
  publisher = {MIT Press},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Cambridge, MA, USA},
  month = jul,
  year = {2005},
  month_numeric = {7}
}