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

Supernetwork-based efficient mapping of deep learning applications to mixed-precision hardware using model

Hadjer Benmeziane1, Corey Lammie2, Irem Boybat2

  • 1IBM Research Europe, Rüschlikon, Switzerland. hadjer.benmeziane@ibm.com.

Nature Communications
|March 28, 2026
PubMed
Summary

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

Demonstration of transformer-based ALBERT model on a 14nm analog AI inference chip.

Nature communications·2025
Same author

Analogue speech recognition based on physical computing.

Nature·2025
Same author

Training of physical neural networks.

Nature·2025
Same author

The growing memristor industry.

Nature·2025
Same author

The inherent adversarial robustness of analog in-memory computing.

Nature communications·2025
Same author

Rapid learning with phase-change memory-based in-memory computing through learning-to-learn.

Nature communications·2025
Same journal

Interplay between oxygen redox and interfacial stability of Li-rich positive electrodes in sulfide-based all-solid-state batteries.

Nature communications·2026
Same journal

Breaking dependence on melanisation imparts diversity to a dogmatic invasion strategy of phytopathogenic fungi.

Nature communications·2026
Same journal

Hydroxyl-rich nanocavities on perovskite enable nearly barrierless intramolecular hydrogen transfer for nitrate electroreduction to ammonia.

Nature communications·2026
Same journal

Household mobility responses to weather extremes in Kyrgyzstan.

Nature communications·2026
Same journal

Autonomous Motion Vision with Tri-bulk-heterojunctioned Organic Adaptation Transistor.

Nature communications·2026
Same journal

Tissue-adhesive hydrogel optical fiber for peripheral optogenetic neuromodulation.

Nature communications·2026
See all related articles
This summary is machine-generated.

Mixed-Precision Supernetwork optimizes deep learning models for heterogeneous accelerators. This framework enhances energy efficiency and model accuracy by intelligently mapping neural network layers to analog and digital hardware.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Hardware Acceleration

Background:

  • Artificial Intelligence (AI) applications are rapidly expanding, requiring efficient and scalable solutions for real-world deployment.
  • Heterogeneous accelerators, which combine analog and digital components, offer potential for localized and energy-efficient neural network computations.
  • Optimizing performance on these accelerators involves a critical balance between energy efficiency and model accuracy through effective neural network layer mapping.

Purpose of the Study:

  • To introduce Mixed-Precision Supernetwork, a unified framework for training mixed-precision supernetworks.
  • To enable seamless integration of quantized digital layers with analog noise-sensitive layers.
  • To develop a mapping-aware adaptation strategy for dynamic layer assignment and hardware-aware architecture search.

Related Experiment Videos

Main Methods:

  • Developed a unified framework for training mixed-precision supernetworks.
  • Incorporated a mapping-aware adaptation strategy for dynamic layer optimization.
  • Utilized hardware-aware architecture search to refine neural network designs.

Main Results:

  • Mixed-Precision Supernetwork achieves mappings approximately 2.2x faster than fully analog approaches.
  • Demonstrates an average increase of 3.4% in model accuracy compared to fully analog methods.
  • Improves energy efficiency by mapping up to 80% of model weights to analog hardware while preserving full-precision accuracy.

Conclusions:

  • Mixed-Precision Supernetwork provides a groundbreaking approach for efficient deep learning model deployment on heterogeneous accelerators.
  • The framework successfully balances energy efficiency and model accuracy through optimized layer mapping.
  • This method enables significant performance gains and enhanced energy efficiency for AI on specialized hardware.