ML-HW Co-Design of Noise-Robust TinyML Models and Always-On Analog Compute-in-Memory Edge Accelerator

Autor: Chuteng Zhou, Fernando Garcia Redondo, Julian Buchel, Irem Boybat, Xavier Timoneda Comas, S. R. Nandakumar, Shidhartha Das, Abu Sebastian, Manuel Le Gallo, Paul N. Whatmough

Publikováno v: IEEE Micro, 42 (6)

Always-on TinyML perception tasks in Internet of Things applications require very high energy efficiency. Analog compute-in-memory (CiM) using nonvolatile memory (NVM) promises high energy efficiency and self-contained on-chip model storage. However,

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e5b41efc4c3d6883ea22e7ee58ea4d59
https://hdl.handle.net/20.500.11850/581240

27.2 M0N0: A Performance-Regulated 0.8-to-38MHz DVFS ARM Cortex-M33 SIMD MCU with 10nW Sleep Power

Autor: Thanusree Achuthan, Jonas Svedas, Graham Knight, Matthew J Walker, Fernando Garcia-Redondo, Anand Savanth, Benoit Labbe, Supreet Jeloka, Pranay Prabhat, Philex Ming-Yan Fan, James Myers

Publikováno v: ISSCC

Recent research has shown subthreshold operation to reduce active energy in low-power MCUs [1], [2], [5]. However, some applications impose additional constraints. For battery-powered sensor nodes deployed in remote locations, the MCU may lie dormant

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::9f40f7d6ab5f3efc31f6cc98c40364d5
https://doi.org/10.1109/isscc19947.2020.9063136

On the Design and Analysis of Reliable RRAM-CMOS Hybrid Circuits

Autor: Fernando Garcia-Redondo, Marisa Lopez-Vallejo

Publikováno v: IEEE Transactions on Nanotechnology. 16:514-522

Resistive switching memories (RRAMs) are one of the most promising alternatives for nonvolatile storage and nonconventional computing systems. However, their behavior, and therefore their reliability, is limited by technology intrinsic constraints. S

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::5d7e8a12116fe6dec50fcbc317976d38
https://doi.org/10.1109/tnano.2017.2697311

A Time-Domain Current-Mode MAC Engine for Analogue Neural Networks in Flexible Electronics

Autor: Pantelis Georgiou, Shidhartha Das, Matthew Douthwaite, Fernando Garcia-Redondo

Publikováno v: 2019 IEEE Biomedical Circuits and Systems Conference (BioCAS)
BioCAS

Flexible electronics is becoming more prevalent in a wide range of applications, particularly wearable biomedical devices. These devices would greatly benefit from in-built intelligence allowing them to process data and identify features, in order to

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0dc4890ad4702f43b614b53bd6c27f72

Applications of Computation-In-Memory Architectures based on Memristive Devices

Autor: Abu Sebastian, Hoang Anh Du Nguyen, Luca Benini, Manuel Le Gallo, Abbas Rahimi, Mottaqiallah Taouil, Fernando Garcia Redondo, Siebren Schaafsma, Said Hamdioui, Sandeep Pande, Shidhartha Das, Geethan Karunaratne, Francky Catthoor

Publikováno v: DATE
2019 Design, Automation & Test in Europe Conference & Exhibition (DATE)
Proceedings of the 2019 Design, Automation and Test in Europe Conference and Exhibition, DATE 2019: Proceedings
Proceedings of the 2019 Design, Automation and Test in Europe Conference and Exhibition, DATE 2019
Proceedings of the 2019 Design, Automation & Test in Europe Conference & Exhibition (DATE)

Today’s computing architectures and device technologies are unable to meet the increasingly stringent demands on energy and performance posed by emerging applications. Therefore, alternative computing architectures are being explored that leverage

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f90b7a782386ad2d46adc25781608ba8
https://doi.org/10.23919/date.2019.8715020