Show simple item record

dc.contributor.authorJokic, Petar, Dr.
dc.contributor.authorAzarkhish, Erfan, Dr.
dc.contributor.authorBonetti, Andrea, Dr.
dc.contributor.authorPons, Marc, Dr.
dc.contributor.authorEmery, Stephane, Dr.
dc.contributor.authorBenini, Luca, Prof. Dr.
dc.identifier.citationACM Transactions on Embedded Computing Systems, Volume 21, Issue 5, September 2022, Article No.: 56, pp 1–36,
dc.description.abstractImplementing embedded neural network processing at the edge requires efficient hardware acceleration that combines high computational throughput with low power consumption. Driven by the rapid evolution of network architectures and their algorithmic features, accelerator designs are constantly being adapted to support the improved functionalities. Hardware designers can refer to a myriad of accelerator implementations in the literature to evaluate and compare hardware design choices. However, the sheer number of publications and their diverse optimization directions hinder an effective assessment. Existing surveys provide an overview of these works but are often limited to system-level and benchmark-specific performance metrics, making it difficult to quantitatively compare the individual effects of each utilized optimization technique. This complicates the evaluation of optimizations for new accelerator designs, slowing-down the research progress. In contrast to previous surveys, this work provides a quantitative overview of neural network accelerator optimization approaches that have been used in recent works and reports their individual effects on edge processing performance. The list of optimizations and their quantitative effects are presented as a construction kit, allowing to assess the design choices for each building block individually. Reported optimizations range from up to 10,000× memory savings to 33× energy reductions, providing chip designers with an overview of design choices for implementing efficient low power neural network accelerators.en_US
dc.description.sponsorshipThis work was supported in part by the Electronic Components and Systems for European Leadership Joint Undertaking ANDANTE under Grant 876925 and in part by the Swiss National Science Foundation (SNSF) BRIDGE under Grant 40B2- 0_181010.en_US
dc.subjectMachine learning, Hardware acceleration, ASIC, Edge Processingen_US
dc.titleA Construction Kit for Efficient Low Power Neural Network Accelerator Designsen_US
dc.type.csemresearchareasData & AIen_US
dc.type.csemresearchareasASICs for the Edgeen_US

Files in this item


This item appears in the following Collection(s)

  • Research Publications
    The “Research Publications” collection provides bibliographic information for scientific papers including conference proceedings and presentations.

Show simple item record