Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials
The Abisko project aims to develop an energy-efficient spiking neural network (SNN) computing architecture and software system capable of autonomous learning and operation. The SNN architecture explores novel neuromorphic devices that are based on resistive-switching materials, such as memristors an...
Published in: | The International Journal of High Performance Computing Applications |
---|---|
Main Authors: | , , , , , , , , , , , , , , , , |
Language: | unknown |
Published: |
2023
|
Subjects: | |
Online Access: | http://www.osti.gov/servlets/purl/1987763 https://www.osti.gov/biblio/1987763 https://doi.org/10.1177/10943420231178537 |
id |
ftosti:oai:osti.gov:1987763 |
---|---|
record_format |
openpolar |
spelling |
ftosti:oai:osti.gov:1987763 2023-07-30T03:55:21+02:00 Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials Vetter, Jeffrey S. Date, Prasanna Fahim, Farah Kulkarni, Shruti R. Maksymovych, Petro Talin, A. Alec Tallada, Marc Gonzalez Vanna-iampikul, Pruek Young, Aaron R. Brooks, David Cao, Yu Gu-Yeon, Wei Lim, Sung Kyu Liu, Frank Marinella, Matthew Sumpter, Bobby Miniskar, Narasinga Rao 2023-07-07 application/pdf http://www.osti.gov/servlets/purl/1987763 https://www.osti.gov/biblio/1987763 https://doi.org/10.1177/10943420231178537 unknown http://www.osti.gov/servlets/purl/1987763 https://www.osti.gov/biblio/1987763 https://doi.org/10.1177/10943420231178537 doi:10.1177/10943420231178537 2023 ftosti https://doi.org/10.1177/10943420231178537 2023-07-11T10:28:04Z The Abisko project aims to develop an energy-efficient spiking neural network (SNN) computing architecture and software system capable of autonomous learning and operation. The SNN architecture explores novel neuromorphic devices that are based on resistive-switching materials, such as memristors and electrochemical RAM. Equally important, Abisko uses a deep codesign approach to pursue this goal by engaging experts from across the entire range of disciplines: materials, devices and circuits, architectures and integration, software, and algorithms. Here, the key objectives of our Abisko project are threefold. First, we are designing an energy-optimized high-performance neuromorphic accelerator based on SNNs. This architecture is being designed as a chiplet that can be deployed in contemporary computer architectures and we are investigating novel neuromorphic materials to improve its design. Second, we are concurrently developing a productive software stack for the neuromorphic accelerator that will also be portable to other architectures, such as field-programmable gate arrays and GPUs. Third, we are creating a new deep codesign methodology and framework for developing clear interfaces, requirements, and metrics between each level of abstraction to enable the system design to be explored and implemented interchangeably with execution, measurement, a model, or simulation. As a motivating application for this codesign effort, we target the use of SNNs for an analog event detector for a high-energy physics sensor. Other/Unknown Material Abisko SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) Abisko ENVELOPE(18.829,18.829,68.349,68.349) The International Journal of High Performance Computing Applications 37 3-4 351 379 |
institution |
Open Polar |
collection |
SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) |
op_collection_id |
ftosti |
language |
unknown |
description |
The Abisko project aims to develop an energy-efficient spiking neural network (SNN) computing architecture and software system capable of autonomous learning and operation. The SNN architecture explores novel neuromorphic devices that are based on resistive-switching materials, such as memristors and electrochemical RAM. Equally important, Abisko uses a deep codesign approach to pursue this goal by engaging experts from across the entire range of disciplines: materials, devices and circuits, architectures and integration, software, and algorithms. Here, the key objectives of our Abisko project are threefold. First, we are designing an energy-optimized high-performance neuromorphic accelerator based on SNNs. This architecture is being designed as a chiplet that can be deployed in contemporary computer architectures and we are investigating novel neuromorphic materials to improve its design. Second, we are concurrently developing a productive software stack for the neuromorphic accelerator that will also be portable to other architectures, such as field-programmable gate arrays and GPUs. Third, we are creating a new deep codesign methodology and framework for developing clear interfaces, requirements, and metrics between each level of abstraction to enable the system design to be explored and implemented interchangeably with execution, measurement, a model, or simulation. As a motivating application for this codesign effort, we target the use of SNNs for an analog event detector for a high-energy physics sensor. |
author |
Vetter, Jeffrey S. Date, Prasanna Fahim, Farah Kulkarni, Shruti R. Maksymovych, Petro Talin, A. Alec Tallada, Marc Gonzalez Vanna-iampikul, Pruek Young, Aaron R. Brooks, David Cao, Yu Gu-Yeon, Wei Lim, Sung Kyu Liu, Frank Marinella, Matthew Sumpter, Bobby Miniskar, Narasinga Rao |
spellingShingle |
Vetter, Jeffrey S. Date, Prasanna Fahim, Farah Kulkarni, Shruti R. Maksymovych, Petro Talin, A. Alec Tallada, Marc Gonzalez Vanna-iampikul, Pruek Young, Aaron R. Brooks, David Cao, Yu Gu-Yeon, Wei Lim, Sung Kyu Liu, Frank Marinella, Matthew Sumpter, Bobby Miniskar, Narasinga Rao Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
author_facet |
Vetter, Jeffrey S. Date, Prasanna Fahim, Farah Kulkarni, Shruti R. Maksymovych, Petro Talin, A. Alec Tallada, Marc Gonzalez Vanna-iampikul, Pruek Young, Aaron R. Brooks, David Cao, Yu Gu-Yeon, Wei Lim, Sung Kyu Liu, Frank Marinella, Matthew Sumpter, Bobby Miniskar, Narasinga Rao |
author_sort |
Vetter, Jeffrey S. |
title |
Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
title_short |
Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
title_full |
Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
title_fullStr |
Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
title_full_unstemmed |
Abisko: Deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
title_sort |
abisko: deep codesign of an architecture for spiking neural networks using novel neuromorphic materials |
publishDate |
2023 |
url |
http://www.osti.gov/servlets/purl/1987763 https://www.osti.gov/biblio/1987763 https://doi.org/10.1177/10943420231178537 |
long_lat |
ENVELOPE(18.829,18.829,68.349,68.349) |
geographic |
Abisko |
geographic_facet |
Abisko |
genre |
Abisko |
genre_facet |
Abisko |
op_relation |
http://www.osti.gov/servlets/purl/1987763 https://www.osti.gov/biblio/1987763 https://doi.org/10.1177/10943420231178537 doi:10.1177/10943420231178537 |
op_doi |
https://doi.org/10.1177/10943420231178537 |
container_title |
The International Journal of High Performance Computing Applications |
container_volume |
37 |
container_issue |
3-4 |
container_start_page |
351 |
op_container_end_page |
379 |
_version_ |
1772811691248058368 |