paper-with-me

Papers

Sparse Periodic Systolic Dataflow for Lowering Latency and Power Dissipation of Convolutional Neural Network Accelerators

2022-06-30 · Jung Hwan Heo, Arash Fayyazi, Amirhossein Esmaili, Massoud Pedram

This paper introduces the sparse periodic systolic (SPS) dataflow, which advances the state-of-the-art hardware accelerator for supporting lightweight neural networks. Specifically, the SPS dataflow enables a novel hardware design approach unlocked by an emergent pruning scheme, periodic pattern-based sparsity (PPS). By exploiting the regularity of PPS, our sparsity-aware compiler optimally reorders the weights and uses a simple indexing unit in hardware to create matches between the weights and activations. Through the compiler-hardware codesign, SPS dataflow enjoys higher degrees of parallelism while being free of the high indexing overhead and without model accuracy loss. Evaluated on popular benchmarks such as VGG and ResNet, the SPS dataflow and accompanying neural network compiler outperform prior work in convolutional neural network (CNN) accelerator designs targeting FPGA devices. Against other sparsity-supporting weight storage formats, SPS results in 4.49x energy efficiency gain while lowering storage requirements by 3.67x for total weight storage (non-pruned weights plus indexing) and 22,044x for indexing memory.

📄 PDF Abstract BibTeX arXiv:2207.00068

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Pruning 설명 없음
Average Pooling 설명 없음
Batch Normalization 설명 없음
ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 설명 없음
Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…

Similar Papers 제목 키워드 기반

FuSeConv: Fully Separable Convolutions for Fast Inference on Systolic Arrays

2021-05-27 · Surya Selvam, Vinod Ganesan, Pratyush Kumar

Both efficient neural networks and hardware accelerators are being explored to speed up DNN inference on edge devices. For example, MobileNet uses depthwise separable convolution to achieve much lower latency, while syst…

Neural Architecture Search

Design and Scaffolded Training of an Efficient DNN Operator for Computer Vision on the Edge

2021-08-25 · Vinod Ganesan, Pratyush Kumar

Massively parallel systolic arrays and resource-efficient depthwise separable convolutions are two promising techniques to accelerate DNN inference on the edge. Interestingly, their combination is inefficient: Computatio…

Neural Architecture Search

DRACO: Co-Optimizing Hardware Utilization, and Performance of DNNs on Systolic Accelerator

2020-06-26 · Nandan Kumar Jha, Shreyas Ravishankar, Sparsh Mittal, Arvind Kaushik 외

The number of processing elements (PEs) in a fixed-sized systolic accelerator is well matched for large and compute-bound DNNs; whereas, memory-bound DNNs suffer from PE underutilization and fail to achieve peak performa…

Computational Efficiency

Hybrid Systolic Array Accelerator with Optimized Dataflow for Edge Large Language Model Inference

2025-07-11 · Chun-Ting Chen, HanGyeol Mun, Jian Meng, Mohamed S. Abdelfattah 외 arxiv

Edge inference for large language models (LLM) offers secure, low-latency, and cost-effective inference solutions. We emphasize that an edge accelerator should achieve high area efficiency and minimize external memory ac…

S2Engine: A Novel Systolic Architecture for Sparse Convolutional Neural Networks

2021-06-15 · Jianlei Yang, Wenzhi Fu, Xingzhou Cheng, Xucheng Ye 외

Convolutional neural networks (CNNs) have achieved great success in performing cognitive tasks. However, execution of CNNs requires a large amount of computing resources and generates heavy memory traffic, which imposes …