Sep 2026· Coins· pp. 1-7· 0 citations· 30 references
TL;DR
A hardware-algorithm co-design approach in which a semi-structured pruning method and a sparsity-aware systolic accelerator are developed jointly and motivated by each other, thereby reducing unnecessary computation and energy consumption.
Abstract
Deep neural networks (DNNs) are increasingly deployed in real-time applications such as autonomous driving, robotics, and computer vision, where high accuracy must be achieved under strict performance and energy constraints. These networks rely heavily on matrix multiplication operations, which impose significant computational and power demands during both training and inference. Moreover, many network parameters become zero as a result of pruning or sparsity-inducing training methods, leading to sparse matrices in which conventional hardware performs a large number of redundant multiplications and additions involving zero-valued operands. This paper presents a hardware-algorithm co-design approach in which a semi-structured pruning method and a sparsity-aware systolic accelerator are developed jointly and motivated by each other. The accelerator is based on a systolic data flow architecture configurable for matrix multiplication operations. Its processing elements (PEs) are designed to bypass zero-valued multiplications and accumulations, thereby reducing unnecessary computation and energy consumption. To further enhance zero-skipping efficiency, we perform a statistical analysis of zero distributions across different network layers. In addition, each PE employs coarse-grained execution by performing multiple multiply–accumulate (MAC) operations per cycle, which, in combination with sparsity-aware execution, improves overall throughput by eliminating redundant operations. Simulation results show that the proposed accelerator reduces the power consumption of matrix multiplication computations with 50% sparsity by 11%. Additionally, the accelerator improves the computation speed for 50% sparse matrices by 100%, while the accuracy degradation of AlexNet, considering pruning from the third layer onward, is only 6%.
The domain-specific hardware accelerators for deep neural network (DNN) inference have been widely adopted, particularly in resource-constrained edge computing scenarios but face a fundamental tradeoff among accuracy, on-chip resource utilization, and inference latency. Traditional compression techniques, such as pruni...
Chung-Mou Pan, Swarnalata Panigrahy, Salma Elmalaki et al.· IEEE Embedded Systems Letter...· 0 citations
Sparse neural networks facilitate accelerated computation and reduced energy consumption by minimizing the total number of operations required. However, conventional digital and analog computing platforms typically yield only modest performance gains when executing sparse models due to hardware primitives that remain u...
Arjun Iyer, Abishek Tyagi, Liam Young et al.· Emerging Topics in Artificia...· 0 citations
Training neural networks in reduced precision can yield significant efficiency gains, yet all existing methods that scale to standard computer vision benchmarks retain floating-point in at least one stage, typically the weight update or optimizer state. We present a training algorithm that scales to modern deep neural...
Hazem Hesham Yousef Shalby, Fabrizio Pittorino, Manuel Roveri· 0 citations
This paper proposes a metrics-driven predictive methodology for FPGA-based DNN acceleration using an OpenCL-based high-level synthesis framework that validates how predictive modeling can help guide FPGA design choices in a way that improves both efficiency and portability for signal and image-processing-oriented DNN a...
Soufien Gdaim, Hachem Bensalem, Ghyslain Gagnon· Journal of Signal Processing...· 0 citations
The contemporary challenges facing computer engineering have led to a focus on improving the efficiency of hardware and computing systems, as well as on optimising the operation of devices for effective data processing. In most modern systems of artificial intelligence, computer vision and digital signal processing, ma...
Y. Klyatchenko, O. Tarasenko-Klyatchenko· International Journal of Eng...· 0 citations
The high and balanced efficiencies prove that learning can be unified with high-performance inference without efficiency compromise for adaptive edge artificial intelligence (AI).
Yue Zuo, N. Ning, Zhi-Qiang Su et al.· IEEE Transactions on Very La...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.