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

A probabilistic active support vector learning algorithm.

Pabitra Mitra1, C A Murthy, Sankar K Pal

  • 1Machine Intelligence Unit, Indian Statistical Institute, Calcutta 700 108, India. pabitra_r@isical.ac.in

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 21, 2004
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

Self Supervised Prediction of Genetic Associations in Comorbid Diseases With Masked Autoencoder Using Hypergraph Representations.

IEEE transactions on computational biology and bioinformatics·2025
Same author

Mental health problems raise the odds of cognitive impairment in COVID-19 survivors.

Frontiers in psychiatry·2024
Same author

Molecular histopathology of matrix proteins through autofluorescence super-resolution microscopy.

Scientific reports·2024
Same author

Prediction and assessment of the impact of COVID-19 lockdown on air quality over Kolkata: a deep transfer learning approach.

Environmental monitoring and assessment·2022
Same author

Detection of Neurogenic Voice Disorders Using the Fisher Vector Representation of Cepstral Features.

Journal of voice : official journal of the Voice Foundation·2022
Same author

Z-number-based AQI in rough set theoretic framework for interpretation of air quality for different thresholds of PM<sub>2.5</sub> and PM<sub>10</sub>.

Environmental monitoring and assessment·2022
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

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

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

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

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

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

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

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

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

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

Achieving Text-based Person Retrieval with Any Granularity.

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

This study introduces a new probabilistic active learning strategy for Support Vector Machines (SVMs) in large datasets. The adaptive confidence factor improves SVM learning efficiency and robustness.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Data Science

Background:

  • Support Vector Machines (SVMs) are powerful classification tools, but their application in large datasets presents computational challenges.
  • Existing active learning strategies for SVMs often focus on points near the separating hyperplane, potentially limiting efficiency.
  • The statistical query model provides a theoretical foundation for understanding learning processes.

Purpose of the Study:

  • To develop a novel probabilistic active learning strategy for SVM design tailored for large-scale data applications.
  • To enhance the robustness and efficiency of SVM learning through a new querying approach.
  • To improve generalization performance, reduce query complexity, and decrease training time in SVM models.

Main Methods:

Related Experiment Videos

  • A probabilistic active learning strategy is proposed, departing from proximity-based querying.
  • The strategy utilizes a distribution determined by the separating hyperplane and an adaptive confidence factor.
  • The adaptive confidence factor is estimated using local information via the k-nearest neighbor principle.

Main Results:

  • The proposed method demonstrates improved generalization performance on real-life datasets.
  • The strategy effectively reduces query complexity compared to existing methods.
  • Significant improvements in training time were observed, highlighting enhanced efficiency.

Conclusions:

  • The developed probabilistic active learning strategy offers a more robust and efficient approach for SVM design in large data scenarios.
  • The adaptive confidence factor, estimated using local information, is key to the method's success.
  • The findings suggest practical benefits for applying this method to real-world large-scale machine learning problems.