DIMCA: An Area-Efficient Digital In-Memory Computing Macro Featuring Approximate Arithmetic Hardware in 28 nm

Recent SRAM-based in-memory computing (IMC) hardware demonstrates high energy efficiency and throughput for matrix-vector multiplication (MVM), the dominant kernel for deep neural networks (DNNs). Earlier IMC macros have employed analog-mixed-signal (AMS) arithmetic hardware. However, those so-calle...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE journal of solid-state circuits Jg. 59; H. 3; S. 960 - 971
Hauptverfasser: Lin, Chuan-Tung, Wang, Dewei, Zhang, Bo, Chen, Gregory K., Knag, Phil C., Krishnamurthy, Ram Kumar, Seok, Mingoo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.03.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Accuracy
Approximate computing
Approximation
approximation-aware training
Arithmetic
Artificial neural networks
Compressors
Computation
Deep learning
Hardware
In-memory computing
in-memory computing (IMC)
Integrated circuit modeling
Mathematical analysis
Multiplication
neural network accelerators
Static random access memory
ISSN:0018-9200, 1558-173X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Recent SRAM-based in-memory computing (IMC) hardware demonstrates high energy efficiency and throughput for matrix-vector multiplication (MVM), the dominant kernel for deep neural networks (DNNs). Earlier IMC macros have employed analog-mixed-signal (AMS) arithmetic hardware. However, those so-called AIMCs suffer from process, voltage, and temperature (PVT) variations. Digital IMC (DIMC) macros, on the other hand, exhibit better robustness against PVT variations, but they tend to require more silicon area. This article proposes novel DIMC hardware featuring approximate arithmetic (DIMCA) to improve area efficiency without hurting compute density (CD). We also propose an approximation-aware training model and a customized number format to compensate for the accuracy degradation caused by the approximation hardware. We prototyped the test chip in 28-nm CMOS. It contains two versions: the DIMCA with single-approximate hardware (DIMCA1) and DIMCA with double-approximate hardware (DIMCA2). The measurement results show that DIMCA1 supports a 4 b-activation and 1 b-weight (4 b/1 b) CNN model, achieving 327 kb/mm2, 458-990 TOPS/W (normalized to 1 b/1 b), 8.27-392 TOPS/mm2 (normalized to 1 b/1 b), and 90.41% accuracy for CIFAR-10. DIMCA2 supports a 1 b/1 b CNN model, achieving 485 kb/mm2, 932-2219 TOPS/W, 14.4-607 TOPS/mm2, and 86.96% accuracy for CIFAR-10.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9200
1558-173X
DOI:10.1109/JSSC.2023.3313519