Retrospective: A CORDIC Based Configurable Activation Function for NN Applications

Omkar Kokane, Gopal Raut, Salim Ullah, Mukul Lokhande, Adam Teman, Akash Kumar, Santosh Kumar Vishvakarma

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A CORDIC-based configuration for the design of Activation Functions (AF) was previously suggested to accelerate ASIC hardware design for resource-constrained systems by providing functional reconfigurability. Since its introduction, this new approach for neural network acceleration has gained widespread popularity, influencing numerous designs for activation functions in both academic and commercial AI processors. In this retrospective analysis, we explore the foundational aspects of this initiative, summarize key developments over recent years, and introduce the DA-VINCI AF tailored for the evolving needs of AI applications. This new generation of dynamically configurable and precision-adjustable activation function cores promise greater adaptability for a range of activation functions in AI workloads, including Swish, SoftMax, SeLU, and GeLU, utilizing the Shift-and-Add CORDIC technique. The previously presented design has been optimized for MAC, Sigmoid, and Tanh functionalities and incorporated into ReLU AFs, culminating in an accumulative NEURIC compute unit. These enhancements position NEURIC as a fundamental component in the resourceefficient vector engine for the realization of AI accelerators that focus on DNNs, RNNs/LSTMs, and Transformers, achieving a quality of results (QoR) of 98.5%.

Original languageEnglish
Title of host publicationIEEE Computer Society Annual Symposium on VLSI, ISVLSI 2025 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)9798331534776
DOIs
StatePublished - 1 Jan 2025
Externally publishedYes
Event28th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2025 - Kalamata, Greece
Duration: 6 Jul 20259 Jul 2025

Publication series

NameProceedings of IEEE Computer Society Annual Symposium on VLSI, ISVLSI
ISSN (Print)2159-3469
ISSN (Electronic)2159-3477

Conference

Conference28th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2025
Country/TerritoryGreece
CityKalamata
Period6/07/259/07/25

Keywords

  • Activation Function
  • AI accelerators
  • CORDIC
  • Reconfigurable Computing
  • Transformers

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Retrospective: A CORDIC Based Configurable Activation Function for NN Applications'. Together they form a unique fingerprint.

Cite this