Lightspeed Binary Neural Networks using Optical Phase-Change Materials

Taha Shahroodi, Rafaela Cardoso, Mahdi Zahedi, Stephan Wong, Alberto Bosio, Ian O'Connor, Said Hamdioui

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review


This paper investigates the potential of a compute-in-memory core based on optical Phase Change Materials (oPCMs) to speed up and reduce the energy consumption of the Matrix-Matrix-Multiplication operation. The paper also proposes a new data mapping for Binary Neural Networks (BNNs) tailored for our oPCM core. The preliminary results show a significant latency improvement irrespective of the evaluated network structure and size. The improvement varies from network to network and goes up to ~1053x.
Original languageEnglish
Title of host publicationProceedings of the 2023 Design, Automation & Test in Europe Conference & Exhibition (DATE)
Place of PublicationPiscataway
Number of pages2
ISBN (Print)979-8-3503-9624-9
Publication statusPublished - 2023
EventDATE 2023: Design, Automation & Test in Europe Conference & Exhibition - Antwerp, Belgium
Duration: 17 Apr 202319 Apr 2023


ConferenceDATE 2023
Internet address

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.


  • Binary Neural Network
  • Optical PCM
  • Computation-In-Memory

Cite this