Conference proceeding
Design Methodology for Embedded Approximate Artificial Neural Networks
GLSVLSI '19 - PROCEEDINGS OF THE 2019 ON GREAT LAKES SYMPOSIUM ON VLSI, pp 489-494
01 Jan 2019
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Artificial neural networks (ANNs) have demonstrated significant promise while implementing recognition and classification applications. The implementation of pre-trained ANNs on embedded systems requires representation of data and design parameters in low-precision fixed-point formats; which often requires retraining of the network. For such implementations, the multiply-accumulate operation is the main reason for resultant high resource and energy requirements. To address these challenges, we present Rox-ANN, a design methodology for implementing ANNs using processing elements (PEs) designed with low-precision fixed-point numbers and high performance and reduced-area approximate multipliers on FPGAs. The trained design parameters of the ANN are analyzed and clustered to optimize the total number of approximate multipliers required in the design. With our methodology, we achieve insignificant loss in application accuracy. We evaluated the design using a LeNet based implementation of the MNIST digit recognition application. The results show a 65.6%, 55.1% and 18.9% reduction in area, energy consumption and latency for a PE using 8-bit precision weights and activations and approximate arithmetic units, when compared to 16-bit full precision, accurate arithmetic PEs.
Metrics
Details
- Title
- Design Methodology for Embedded Approximate Artificial Neural Networks
- Creators
- Adarsha Balaji - Drexel UniversitySalim Ullah - TU DresdenAnup Das - Drexel UniversityAkash Kumar - TU DresdenAssoc Comp Machinery
- Publication Details
- GLSVLSI '19 - PROCEEDINGS OF THE 2019 ON GREAT LAKES SYMPOSIUM ON VLSI, pp 489-494
- Series
- Proceedings - Great Lakes Symposium on VLSI
- Publisher
- Assoc Computing Machinery
- Number of pages
- 6
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Electrical and Computer Engineering
- Web of Science ID
- WOS:000474339800097
- Scopus ID
- 2-s2.0-85083186464
- Other Identifier
- 991019238706204721
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:
- Collaboration types
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Computer Science, Theory & Methods