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

Exploiting application locality to design low-complexity, highly performing, and power-aware embedded classifiers.

Cesare Alippi1, Fabio Scotti

  • 1Dipartimento di Elettronica e Informazione, Politecnico di Milano, 20133 Milano, Italy. alippi@elet.polimi.it

IEEE Transactions on Neural Networks
|May 26, 2006
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

Prescribed-Performance Control of Variable-Order Fractional-Order Nonlinear Systems Through Adaptive Dynamic Programming.

IEEE transactions on cybernetics·2026
Same author

Non-metric traits of the cranium: are they related one with each other? A novel approach to skeletal variants.

International journal of legal medicine·2026
Same author

FX-DARTS: Designing Topology-Unconstrained Architectures With Differentiable Architecture Search and Entropy-BasedSuper-Network Shrinking.

IEEE transactions on neural networks and learning systems·2025
Same author

From traditional to innovative: implications of cranial non-metric traits in personal identification.

International journal of legal medicine·2025
Same author

A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection.

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

Dynamically aggregating MLPs and CNNs for skin lesion segmentation with geometry regularization.

Computer methods and programs in biomedicine·2023
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Embedded classifiers leverage input locality for efficient design. This approach reduces computational complexity and power consumption while maintaining high accuracy in machine learning models.

Area of Science:

  • Computer Science
  • Machine Learning
  • Embedded Systems

Background:

  • Temporal and spatial locality in input data is a key property for designing efficient embedded classifiers.
  • Existing methods may not fully exploit this locality, leading to higher computational demands.

Purpose of the Study:

  • To investigate the translation of input data locality into the design of embedded classifiers.
  • To develop computational complexity and power-aware classifier designs suitable for implementation.

Main Methods:

  • Utilizing classifiers from the gated-parallel family, known for their suitability in exploiting locality.
  • Implementing a master-enabling module to control and activate only one subclassifier at a time, switching others off.

Main Results:

Related Experiment Videos

  • Classifiers designed with locality properties achieved accuracy comparable to those without.
  • A significant reduction in computational complexity and power consumption was observed.

Conclusions:

  • Exploiting temporal and spatial locality is an effective strategy for designing efficient embedded classifiers.
  • Gated-parallel classifiers offer a suitable architecture for leveraging locality, leading to power and complexity savings.