Faster Convolution Inference Through Using Pre-Calculated Lookup Tables

04/04/2021
by   Grigor Gatchev, et al.
0

Low-cardinality activations permit an algorithm based on fetching the inference values from pre-calculated lookup tables instead of calculating them every time. This algorithm can have extensions, some of which offer abilities beyond those of the currently used algorithms. It also allows for a simpler and more effective CNN-specialized hardware.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/27/2013

An Odds Ratio Based Inference Engine

Expert systems applications that involve uncertain inference can be repr...
research
05/19/2023

Computational models of sound-quality metrics using method for calculating loudness with gammatone/gammachirp auditory filterbank

Sound-quality metrics (SQMs), such as sharpness, roughness, and fluctuat...
research
08/30/2018

Minimal inference from incomplete 2x2-tables

Estimates based on 2x2 tables of frequencies are widely used in statisti...
research
04/16/2022

Logical Inference for Counting on Semi-structured Tables

Recently, the Natural Language Inference (NLI) task has been studied for...
research
10/16/2020

A Conglomerate of Multiple OCR Table Detection and Extraction

Information representation as tables are compact and concise method that...
research
10/06/2019

Clustering case statements for indirect branch predictors

We present an O(nlogn) algorithm to compile a switch statement into jump...
research
01/15/2013

Recurrent Online Clustering as a Spatio-Temporal Feature Extractor in DeSTIN

This paper presents a basic enhancement to the DeSTIN deep learning arch...

Please sign up or login with your details

Forgot password? Click here to reset