Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Principal surfaces from unsupervised kernel regression.

Peter Meinicke1, Stefan Klanke, Roland Memisevic

  • 1P. Meinicke is with the Bioinformatics Department, Faculty of Biology, University of Göttingen, Goldschmidtstr. 1, 37077 Göttingen, Germany. pmeinic@gwdg.de

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 22, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

CoCoPyE: feature engineering for learning and prediction of genome quality indices.

GigaScience·2024
Same author

Comparative analysis of motor skill acquisition in a novel bimanual task: the role of mental representation and sensorimotor feedback.

Frontiers in human neuroscience·2024
Same author

Exploring motor skill acquisition in bimanual coordination: insights from navigating a novel maze task.

Scientific reports·2024
Same author

Face Generation and Editing With StyleGAN: A Survey.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

Perception and simulation during concept learning.

Psychological review·2023
Same author

Action Selection and Execution in Everyday Activities: A Cognitive Robotics and Situation Model Perspective.

Topics in cognitive science·2021
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a new nonparametric method for learning principal surfaces using kernel regression. The approach simplifies model selection and allows fitting surfaces in diverse feature spaces.

Area of Science:

  • Machine Learning
  • Data Analysis
  • Dimensionality Reduction

Background:

  • Principal surfaces are essential for understanding complex data structures.
  • Existing methods for learning principal surfaces often face challenges in model selection and feature space adaptability.
  • Nonlinear spectral methods and principal component analysis (PCA) are common initialization techniques.

Purpose of the Study:

  • To develop a novel nonparametric approach for learning principal surfaces.
  • To address limitations in model selection and computational cost of existing methods.
  • To enable the fitting of principal surfaces in general feature spaces.

Main Methods:

  • Utilizing an unsupervised formulation of the Nadaraya-Watson kernel regression estimator.

Related Experiment Videos

  • Employing leave-one-out cross-validation for parameter estimation and model selection.
  • Incorporating nonlinear spectral methods for parameter initialization.

    • Main Results:

    • The proposed method offers a practical solution to the model selection problem without extra computational cost.
    • It allows for convenient initialization beyond traditional linear PCA.
    • The approach successfully fits principal surfaces in general feature spaces, demonstrated on simulated and real data.

    Conclusions:

    • The new nonparametric method provides an efficient and flexible tool for principal surface learning.
    • It overcomes key limitations of previous approaches, enhancing applicability in diverse data scenarios.
    • The technique is validated through experimental results on various datasets.