Redirigiendo al acceso original de articulo en 17 segundos...
Inicio  /  Algorithms  /  Vol: 15 Par: 7 (2022)  /  Artículo
ARTÍCULO
TITULO

Linear Computation Coding: A Framework for Joint Quantization and Computing

Ralf Reiner Müller    
Bernhard Martin Wilhelm Gäde and Ali Bereyhi    

Resumen

Here we introduce the new concept of computation coding. Similar to how rate-distortion theory is concerned with the lossy compression of data, computation coding deals with the lossy computation of functions. Particularizing to linear functions, we present an algorithmic approach to reduce the computational cost of multiplying a constant matrix with a variable vector, which requires neither a matrix nor vector having any particular structure or statistical properties. The algorithm decomposes the constant matrix into the product of codebook and wiring matrices whose entries are either zero or signed integer powers of two. For a typical application like the implementation of a deep neural network, the proposed algorithm reduces the number of required addition units several times. To achieve the accuracy of 16-bit signed integer arithmetic for 4k-vectors, no multipliers and only 1.5 adders per matrix entry are needed.

 Artículos similares

       
 
Hongming Li, Lilai Zhang, Hao Cao and Yirui Wu    
Deoxyribonucleic Acid (DNA) computing has demonstrated great potential in data encryption due to its capability of parallel computation, minimal storage requirement, and unbreakable cryptography. Focusing on high-dimensional image data for encryption wit... ver más
Revista: Applied Sciences

 
Byoung-Gyu Song, Jong-Jin Bae and Namcheol Kang    
We investigated the stochastic response of a person sitting in a driving vehicle to quantify the impact of an uncertain parameter important in controlling defect reduction in terms of ride comfort. Using CarSim software and MATLAB/Simulink, we developed ... ver más
Revista: Applied Sciences

 
Luigi Rovito, Andrea De Lorenzo and Luca Manzoni    
Stream ciphers usually rely on highly secure Boolean functions to ensure safe communication within unsafe channels. However, discovering secure Boolean functions is a non-trivial optimization problem that has been addressed by many optimization technique... ver más
Revista: Algorithms

 
P. Ashok and B. Bala Tripura Sundari    
Stochastic circuits are used in applications that require low area and power consumption. The computing performed using these circuits is referred to as Stochastic computing (SC). The arithmetic operations in this computing can be realized using minimum ... ver más
Revista: Computation

 
Alexandr Kuznetsov, Mikolaj Karpinski, Ruslana Ziubina, Sergey Kandiy, Emanuele Frontoni, Oleksandr Peliukh, Olga Veselska and Ruslan Kozak    
The problem of nonlinear substitution generation (S-boxes) is investigated in many related works in symmetric key cryptography. In particular, the strength of symmetric ciphers to linear cryptanalysis is directly related to the nonlinearity of substituti... ver más
Revista: Information