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

Efficient adaptive learning for classification tasks with binary units

J M Torres Moreno1, M B Gordon

  • 1Departement de Recherche Fondamentale sur la Matière Condensée, CEA Grenoble, Grenoble, France.

Neural Computation
|June 6, 1998
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

Correction to: Oral octreotide capsules for the treatment of acromegaly: comparison of 2 phase 3 trial results.

Pituitary·2021
Same author

Oral octreotide capsules for the treatment of acromegaly: comparison of 2 phase 3 trial results.

Pituitary·2021
Same author

mRNA of luteal genes associated with progesterone synthesis, maintenance, and apoptosis in dairy heifers and lactating dairy cows.

Animal reproduction science·2010
Same author

Pregnancy rates to timed artificial insemination in dairy cows treated with gonadotropin-releasing hormone or porcine luteinizing hormone.

Theriogenology·2009
Same author

Variations in adequate levothyroxine replacement therapy in patients with different causes of hypothyroidism.

Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists·2004
Same author

SpikeCell: a deterministic spiking neuron.

Neural networks : the official journal of the International Neural Network Society·2003
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

NetLines, a novel incremental learning algorithm, creates compact feedforward neural networks for classification. This method ensures finite hidden units for binary and real-valued inputs, outperforming existing benchmarks.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computer Science

Background:

  • Incremental learning algorithms are crucial for efficient classification tasks.
  • Developing compact neural networks with guaranteed convergence is a key challenge.
  • Existing methods may struggle with both binary and real-valued input patterns.

Purpose of the Study:

  • Introduce NetLines, a new incremental learning algorithm for classification.
  • Demonstrate its ability to generate small, compact feedforward neural networks.
  • Validate its performance against established benchmarks for generalization and network size.

Main Methods:

  • Developed an incremental learning algorithm (NetLines) for binary and real-valued inputs.
  • Designed networks with a single hidden layer of binary units and binary outputs.

Related Experiment Videos

  • Proved a convergence theorem guaranteeing finite hidden units.
  • Implemented a multi-class extension applicable to any binary classifier.

    • Main Results:

    • NetLines generates small, compact feedforward neural networks.
    • A convergence theorem confirms the existence of finite hidden units.
    • Performance on benchmarks shows competitive generalization error and network size.
    • Early stopping reduces overfitting but does not enhance generalization.

    Conclusions:

    • NetLines is an effective incremental learning algorithm for classification.
    • The algorithm produces efficient and compact neural network models.
    • It offers a viable solution for both binary and multi-class problems with real-valued inputs.