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

Fast neural networks without multipliers.

M Marchesi1, G Orlandi, F Piazza

  • 1Dipartimento di Ingegneria Biofisica e Elettron., Genova Univ.

IEEE Transactions on Neural Networks
|January 1, 1993
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Neural Circuits01:25

Neural Circuits

Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Multi-input and Multi-variable systems01:22

Multi-input and Multi-variable systems

Cruise control systems in cars are designed as multi-input systems to maintain a driver's desired speed while compensating for external disturbances such as changes in terrain. The block diagram for a cruise control system typically includes two main inputs: the desired speed set by the driver and any external disturbances, such as the incline of the road. By adjusting the engine throttle, the system maintains the vehicle's speed as close to the desired value as possible.
In the absence of...

You might also read

Related Articles

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

Sort by
Same author

Chimeric antigen receptor T cell therapy cardiotoxicity: a narrative review.

Cardio-oncology (London, England)·2026
Same author

Evidence for the Collective Nature of Radial Flow in Pb+Pb Collisions with the ATLAS Detector.

Physical review letters·2026
Same author

Evidence for the Dimuon Decay of the Higgs Boson in pp Collisions with the ATLAS Detector.

Physical review letters·2025
Same author

Evidence for Longitudinally Polarized W Bosons in the Electroweak Production of Same-Sign W Boson Pairs in Association with Two Jets in pp Collisions at sqrt[s]=13  TeV with the ATLAS Detector.

Physical review letters·2025
Same author

Isolated shoulder palsy due to cortical infarction: localisation and electrophysiological correlates of recovery.

BMJ case reports·2025
Same author

Observation of tt[over ¯] Production in Pb+Pb Collisions at sqrt[s_{NN}]=5.02  TeV with the ATLAS Detector.

Physical review letters·2025

This study introduces efficient Multilayer Perceptrons (MLPs) using power-of-two weights, reducing hardware needs. A backpropagation learning method is presented for these simplified neural networks, proving effective in pattern recognition tasks.

Area of Science:

  • Artificial Intelligence
  • Computer Engineering
  • Machine Learning

Background:

  • Traditional Multilayer Perceptrons (MLPs) rely on complex multiplication operations.
  • Digital implementation of MLPs often faces constraints in chip area and computation time.

Purpose of the Study:

  • To introduce a novel MLP architecture with weights restricted to powers of two or sums of powers of two.
  • To develop an efficient digital implementation and a compatible learning procedure for these specialized MLPs.

Main Methods:

  • Developed MLPs with weights constrained to powers of two or sums of powers of two.
  • Utilized shift registers instead of multipliers for forward-mode computation in digital implementations.
  • Adapted a backpropagation algorithm for training these networks, requiring offline full real arithmetic.

Related Experiment Videos

Main Results:

  • Demonstrated significant savings in chip area and computation time due to the elimination of multipliers.
  • Validated the proposed method through pattern recognition test cases with varying MLP hidden layer sizes.
  • Confirmed the validity and generalization capability of the trained networks.

Conclusions:

  • The proposed MLP architecture offers a practical and efficient digital implementation.
  • The backpropagation-based learning procedure is effective for training these specialized neural networks.
  • The method shows promise for applications requiring reduced computational complexity and hardware resources.