22.1 A 12.4TOPS/W @ 136GOPS AI-IoT System-on-Chip with 16 RISC-V, 2-to-8b Precision-Scalable DNN Acceleration and 30%-Boost Adaptive Body Biasing

Conti, Francesco; Rossi, Davide; Paulin, Gianna; Garofalo, Anaelo; Mauro, Alfio Di; Rutishauer, Georg; Ottavi, Gian marco; Eggimann, Manuel; Okuhara, Hayate; Huard, V.; Montfort, O.; Jure, Lionel; Exibard, N.; Gouedo, Pascal; Louvat, Mathieu; Botte, Emmanuel; Benini, Luca

doi:10.1109/isscc42615.2023.10067643

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2023

2024

Publication Types

Select...

Article3

Other2

Relationship

Self Cite0

Independent5

Authors

Journals

Cited by 17 publications

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

Radiation Effects in VLSI Circuits - Part II: Hardening Techniques

Kannaujiya,

Shah

2024

IETE Technical Review

View full text Add to dashboard Cite

Radiation Effects in VLSI Circuits - Part II: Hardening Techniques

Kannaujiya,

Shah

2024

IETE Technical Review

View full text Add to dashboard Cite

Pianissimo: A Sub-mW Class DNN Accelerator With Progressively Adjustable Bit-Precision

Suzuki,

Yu,

Yasunaga

et al. 2024

IEEE Access

View full text Add to dashboard Cite

With the widespread adoption of edge AI, the diversity of application requirements and fluctuating computational demands present significant challenges. Conventional accelerators suffer from increased memory footprints due to the need for multiple models to adapt to these varied requirements over time. In such dynamic edge conditions, it is crucial to accommodate these changing computational needs within strict memory and power constraints while maintaining the flexibility to support a wide range of applications. In response to these challenges, this article proposes a sub-mW class inference accelerator called Pianissimo that achieves competitive power efficiency while flexibly adapting to changing edge environment conditions at the architecture level. The heart of the design concept is a novel datapath architecture with a progressive bit-by-bit datapath. This unique datapath is augmented by software-hardware (SW-HW) cooperative control with a reduced instruction set computer processor and HW counters. The integrated SW-HW control enables adaptive inference schemes of adaptive/mixed precision and Block Skip, optimizing the balance between computational efficiency and accuracy. The 40 nm chip, with 1104 KB memory, dissipates 793-1032 µW at 0.7 V on MobileNetV1, achieving 0.49-1.25 TOPS/W at this ultra-low power range.INDEX TERMS Ultra-low power, sub-mW, progressive bit-serial datapath, bit-scalable accelerators, adaptive inference, neural network, SW-HW cooperative control.

show abstract