Principles of Nonparametric Learning
Invited Lecturers
- Nicolò Cesa Bianchi (Università di Milano, Crema, Italy)
- 5 lectures on: On-line learning, universal prediction, universal coding, universal portfolio selection.
- Luc Devroye (McGill University, Montreal, Canada)
- 5 lectures on: Nonparametric density estimation, automatic optimal selection of density estimates, L1-theory, the kernel estimate, histogram, series, and wavelet estimates.
- Laszlo Gyorfi (Budapest Univ. of Techn. & Ec., Budapest, Hungary)
- 5 lectures on: Nonparametric pattern recognition, universal consistency, local majority rules (partitioning, kernel, nearest neighbor), regression function estimation, local averaging estimates (partitioning, kernel, nearest neighbor).
- Michael Kohler (Universitat Stuttgart, Stuttgart, Germany)
- 5 lectures on: Regression function estimation, empirical risk minimization, spline estimates, penalized least squares estimates, neural networks, applications in data mining.
- Tamas Linder (QueenO~s University at Kingston, Kingston, Ontario, Canada)
- 5 lectures on: Principles of lossy data compression, vector quantization, universal quantization, clustering, optimal empirical vector quantization.
- Gabor Lugosi (Pompeu Fabra University, Barcelona, Spain)
- 5 lectures on: Statistical learning theory, pattern recognition, empirical risk minimization, Vapnik-Chervonenkis theory, complexity regularization, error estimation, support vector machines.
- Michèle Sebag (Lab. Mec. des Solides, Palaiseau cedex, France)
- 5 lectures on: Evolutionary computation, genetic programming, nonparametric learning with background knowledge, co-evolution, identification of macro-mechanical models.