Simplicial Piecewise Linear Computation Complexity for Vector-Vector Products

Research output: Conference proceeding/Chapter in Book/Report/Conference Paperpeer-review

Abstract

In this paper we analyze several simplicial piecewise-linear digital computation architectures that are good candidates to replace the multiply and accumulate (MAC) units in deep neural networks. We analyze them in terms of area and energy of a digital VLSI realization and compare the results with the classic MAC architectures.
Original languageEnglish
Title of host publication2020 Argentine Conference on Electronics (CAE)
Pages108-113
Number of pages6
DOIs
Publication statusPublished - 28 Feb 2020
Externally publishedYes
Event2020 Argentine Conference on Electronics (CAE) - Buenos Aires, Argentina
Duration: 27 Feb 202028 Feb 2020

Publication series

Name2020 Argentine Conference on Electronics (CAE)

Conference

Conference2020 Argentine Conference on Electronics (CAE)
Period27/02/2028/02/20

Keywords

  • Linear combination
  • piecewise-linear functions
  • vector-vector multiplication

Fingerprint

Dive into the research topics of 'Simplicial Piecewise Linear Computation Complexity for Vector-Vector Products'. Together they form a unique fingerprint.

Cite this