Files and links (1)
SubmittedOpen Access (License Unspecified), Open
Conference proceeding
A Design Methodology for Fault-Tolerant Computing using Astrocyte Neural Networks
PROCEEDINGS OF THE 19TH ACM INTERNATIONAL CONFERENCE ON COMPUTING FRONTIERS 2022 (CF 2022), pp 169-172
01 Jan 2022
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
We propose a design methodology to facilitate fault tolerance of deep learning models. First, we implement a many-core fault-tolerant neuromorphic hardware design, where neuron and synapse circuitries in each neuromorphic core are enclosed with astrocyte circuitries, the star-shaped glial cells of the brain that facilitate self-repair by restoring the spike firing frequency of a failed neuron using a closed-loop retrograde feedback signal. Next, we introduce astrocytes in a deep learning model to achieve the required degree of tolerance to hardware faults. Finally, we use a system software to partition the astrocyte-enabled model into clusters and implement them on the proposed fault-tolerant neuromorphic design. We evaluate this design methodology using seven deep learning inference models and show that it is both area- and power-efficient.
Metrics
Details
- Title
- A Design Methodology for Fault-Tolerant Computing using Astrocyte Neural Networks
- Creators
- Murat Isik - Drexel UniversityAnkita Paul - Drexel UniversityM. Lakshmi Varshika - Drexel UniversityAnup Das - Drexel University
- Publication Details
- PROCEEDINGS OF THE 19TH ACM INTERNATIONAL CONFERENCE ON COMPUTING FRONTIERS 2022 (CF 2022), pp 169-172
- Conference
- CF '22: the 19th ACM International Conference on Computing Frontiers, 19th (Turin Italy, 17 May 2022–22 May 2022)
- Publisher
- Assoc Computing Machinery
- Number of pages
- 4
- Grant note
- CCF-1942697 / National Science Foundation Faculty Early Career Development Award; National Science Foundation (NSF)
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Electrical and Computer Engineering
- Web of Science ID
- WOS:000934076200019
- Scopus ID
- 2-s2.0-85130718495
- Other Identifier
- 991020201846004721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Theory & Methods