计算机科学
现场可编程门阵列
冗余(工程)
粒度
计算机体系结构
推论
前馈
计算机工程
人工神经网络
编码(内存)
可重组计算
计算机硬件
嵌入式系统
过程(计算)
建筑
并行计算
加法器
相容性(地球化学)
钥匙(锁)
前馈神经网络
逻辑综合
设计过程
人工智能
算法设计
深层神经网络
编码(社会科学)
作者
Chenming Zhang,Lei Gong,Chao Wang,Xuehai Zhou
出处
期刊:Proceedings
[Association for Computing Machinery]
日期:2025-06-22
卷期号:: 1-7
被引量:1
标识
DOI:10.1109/dac63849.2025.11133249
摘要
Exploring bit-level sparsity in the MAC process has been proven to be an important method for improving the efficiency of neural network feedforward processing. The reconfigurable platform offers possibilities for identifying the bitlevel unstructured redundancy during inference with different DNN models. Researchers noticed significant progress in valueaware accelerators on ASICs, yet we are concerned about the few studies on FPGAs. This paper observed the limitations of implementing bit-level sparsity optimizations using FPGA and proposed a software/architecture co-design solution. Specifically, by introducing LUT-friendly encoding with adaptable granularity and hardware structure supporting multiplication time uncertainty, we achieved a better trade-off between potential redundancy and accuracy with compatibility and scalability. Experiments show that under accurate calculation, PEs are up to $2.2 \times$ smaller than bit-parallel ones, and our design boosts performance by $1.04 \times$ to $1.74 \times$ and $1.40 \times$ to $2.79 \times$ over bitparallel and Booth-based designs, respectively.
科研通智能强力驱动
Strongly Powered by AbleSci AI