A deterministic approach to stochastic computation

Jenson, Devon; Riedel, Marc

doi:10.1145/2966986.2966988

Cited by 87 publications

(56 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, the nature of randomness in SC could cause some inherent errors in computation. Thus deterministic code [8] of SC has been proposed to produce completely accurate result. Fig.…”

Section: Background and Motivation 21 Backgroundmentioning

confidence: 99%

An Energy-Efficient Mixed-Signal Parallel Multiply-Accumulate (MAC) Engine Based on Stochastic Computing

Zhang

Song

Wang

et al. 2019

2019 IEEE 13th International Conference on ASIC (ASICON)

View full text Add to dashboard Cite

Convolutional neural networks (CNN) have achieved excellent performance on various tasks, but deploying CNN to edge is constrained by the high energy consumption of convolution operation. Stochastic computing (SC) is an attractive paradigm which performs arithmetic operations with simple logic gates and low hardware cost. This paper presents an energy-efficient mixed-signal multiply-accumulate (MAC) engine based on SC. A parallel architecture is adopted in this work to solve the latency problem of SC. The simulation results show that the overall energy consumption of our design is 5.03pJ per 26-input MAC operation under 28nm CMOS technology.

show abstract

“…Besides, the nature of randomness in SC could cause some inherent errors in computation. Thus deterministic code [8] of SC has been proposed to produce completely accurate result. Fig.…”

Section: Background and Motivation 21 Backgroundmentioning

confidence: 99%

An Energy-Efficient Mixed-Signal Parallel Multiply-Accumulate (MAC) Engine Based on Stochastic Computing

Zhang

Song

Wang

et al. 2019

2019 IEEE 13th International Conference on ASIC (ASICON)

View full text Add to dashboard Cite

show abstract

“…Secondly, accuracy is improved by using circuits with correlated inputs (CSC) [6,19], which reduce the area cost. Thirdly, accuracy can be enhanced by performing SC operations using deterministic bit-streams [20,21]. These works obtained accurate arithmetic operations, but the entire bit-stream should be processed to avoid significant truncation error [22].…”

Section: Related Workmentioning

confidence: 99%

Accurate and compact stochastic computations by exploiting correlation

Abdellatef¹,

Khalil-Hani²,

Shaikh-Husin³

2019

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Recent studies have shown, contrary to what was previously believed, that by exploiting correlation in stochastic computing (SC) designs, more accurate SC circuits with low area cost can be realized. However, if these basic SC circuits or blocks are cascaded in series to form a large complex system, correlation between stochastic numbers (SNs) from one block to the next would be lost; thus, inaccuracies are introduced. In this study, we propose correlating circuits to be used in building complex correlated SC systems. One of the circuits is the correlator that restores lost correlations between two SNs due to previous processing. In addition, a correlated SN generator is introduced to generate SN correlated to a specific SN. Experimental results show that our methods have improved the accuracy of stochastic computation and preserved the stochastic computing correlation without the need for conversion from SC to the conventional binary-encoded computing, and vice versa. Consequently, lower latency and lower area cost are achieved.

show abstract

“…Other works [9,14,19,24] have proposed "deterministic bitstreams" (similar to the deterministic bitstreams in this article but not PDM format) to reduce latency. Unfortunately, this work only describes how to implement single or a handful of arithmetic operations and does not discuss how to extend the work to cascaded, generic, dependent, and/or recurrent operations, such as those required for the algorithms explored in this article.…”

Section: Related Work On Deterministic Bitstreamsmentioning

confidence: 99%

B it SAD v2

Daruwalla

Zhuo

Shukla

et al. 2019

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Computer vision and machine learning algorithms operating under a strict power budget require an alternate computing paradigm. While bitstream computing (BC) satisfies these constraints, creating BC systems is difficult. To address the design challenges, we propose compiler extensions to BitSAD, a DSL for BC. Our work enables bit-level software emulation and automated generation of hierarchical hardware, discusses potential optimizations, and proposes compiler phases to implement those optimizations in a hardware-aware manner. Finally, we introduce population coding, a parallelization scheme for stochastic computing that decreases latency without sacrificing accuracy, and provide theoretical and experimental guarantees on its effectiveness.

show abstract

A deterministic approach to stochastic computation

Cited by 87 publications

References 7 publications

An Energy-Efficient Mixed-Signal Parallel Multiply-Accumulate (MAC) Engine Based on Stochastic Computing

An Energy-Efficient Mixed-Signal Parallel Multiply-Accumulate (MAC) Engine Based on Stochastic Computing

Accurate and compact stochastic computations by exploiting correlation

B it SAD v2

Contact Info

Product

Resources

About