首页/论文/1906.03474

arXiv 1906.034742019-06-08cs.LG

5 Parallel Prism: A topology for pipelined implementations of convolutional neural networks using computational memory

IBM Research – Zurich（IBM研究院（苏黎世））；ETH Zurich（苏黎世联邦理工学院）

Martino Dazzi, Abu Sebastian, Pier Andrea Francese, Thomas Parnell, Luca Benini, Evangelos Eleftheriou

英文摘要：

In-memory computing is an emerging computing paradigm that could enable deeplearning inference at significantly higher energy efficiency and reduced latency. The essential idea is to map the synaptic weights corresponding to each layer to one or more computational memory (CM) cores. During inference, these cores perform the associated matrix-vector multiply operations in place with O(1) time complexity, thus obviating the need to move the synaptic weights to an additional processing unit. Moreover, this architecture could enable the execution of these networks in a highly pipelined fashion. However, a key challenge is to design an efficient communication fabric for the CM cores. Here, we present one such communication fabric based on a graph topology that is well suited for the widely successful convolutional neural networks (CNNs). We show that this communication fabric facilitates the pipelined execution of all state of-the-art CNNs by proving the existence of a homomorphism between one graph representation of these networks and the proposed graph topology. We then present a quantitative comparison with established communication topologies and show that our proposed topology achieves the lowest bandwidth requirements per communication channel. Finally, we present a concrete example of mapping ResNet-32 onto an array of CM cores.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Pipelined Training with Stale Weights of Deep Convolutional Neural Networks

University of Toronto（多伦多大学）

The growth in the complexity of Convolutional Neural Networks (CNNs) is increasing interest in partitioning a network across multiple accelerators during training and pipelining the backpropagation computations over the accelerators. Existing approaches avoid or limit the use of stale weights through techniques such as micro-batching or weight stashing. These techniques either underutilize of accelerators or increase memory footprint. We explore the impact of stale weights on the statistical efficiency and performance in a pipelined backpropagation scheme that maximizes accelerator utilization…

2019-12-29cs.DCcs.LG

PipeGCN: Efficient Full-Graph Training of Graph Convolutional Networks with Pipelined Feature Communication

Rice University（莱斯大学）；UIUC（伊利诺伊大学厄巴纳-香槟分校）

Graph Convolutional Networks (GCNs) is the state-of-the-art method for learning graph-structured data, and training large-scale GCNs requires distributed training across multiple accelerators such that each accelerator is able to hold a partitioned subgraph. However, distributed GCN training incurs prohibitive overhead of communicating node features and feature gradients among partitions for every GCN layer during each training iteration, limiting the achievable training efficiency and model scalability. To this end, we propose PipeGCN, a simple yet effective scheme that hides the communicatio…

2022-03-20cs.AIcs.LG

PRISM：通过对称多分辨率卷积层实现轻量级多变量时间序列分类

PRISM: Lightweight Multivariate Time-Series Classification through Symmetric Multi-Resolution Convolutional Layers

ICube, University of Strasbourg（斯特拉斯堡大学ICube实验室）

PRISM通过对称多分辨率卷积层实现轻量级多变量时间序列分类，采用通道无关方式，在早期阶段应用多分辨率对称卷积滤波器，有效减少参数数量并保持完整感受野，优于现有CNN和Transformer模型。

2026-04-04cs.AIcs.LG

Event Arguments Extraction via Dilate Gated Convolutional Neural Network with Enhanced Local Features

National University of Defense Technology（国防科技大学）

Event Extraction plays an important role in information-extraction to understand the world. Event extraction could be split into two subtasks: one is event trigger extraction, the other is event arguments extraction. However, the F-Score of event arguments extraction is much lower than that of event trigger extraction, i.e. in the most recent work, event trigger extraction achieves 80.7%, while event arguments extraction achieves only 58%. In pipelined structures, the difficulty of event arguments extraction lies in its lack of classification feature, and the much higher computation consumptio…

2020-06-02cs.CLcs.LG

LayerPipe2: 通过改进的指数移动平均实现多阶段流水线和权重重计算以训练神经网络

LayerPipe2: Multistage Pipelining and Weight Recompute via Improved Exponential Moving Average for Training Neural Networks

Computer Engineering, University of Minnesota Twin Cities Minneapolis, MN 55455, USA

LayerPipe2通过改进的指数移动平均方法，系统地推导了流水线技术，解决了流水线中梯度延迟的计算问题，并通过流水线感知的移动平均减少存储开销，实现可控的通信计算权衡的可扩展流水线训练。

2025-12-09cs.AIcs.ARcs.LG

Systolic-CNN: An OpenCL-defined Scalable Run-time-flexible FPGA Accelerator Architecture for Accelerating Convolutional Neural Network Inference in Cloud/Edge Computing

Arizona State University（亚利桑那州立大学）

This paper presents Systolic-CNN, an OpenCL-defined scalable, run-time-flexible FPGA accelerator architecture, optimized for accelerating the inference of various convolutional neural networks (CNNs) in multi-tenancy cloud/edge computing. The existing OpenCL-defined FPGA accelerators for CNN inference are insufficient due to limited flexibility for supporting multiple CNN models at run time and poor scalability resulting in underutilized FPGA resources and limited computational parallelism. Systolic-CNN adopts a highly pipelined and paralleled 1-D systolic array architecture, which efficiently…

2020-12-06cs.AIcs.ARcs.LG

更多相关论文收起更多论文

High-Throughput CNN Inference on Embedded ARM big.LITTLE Multi-Core Processors

National University of Singapore（新加坡国立大学）；Indian Institute of Technology Dharwad（印度理工学院达尔瓦德分校）；Indian Institute of Technology Ropar（印度理工学院罗巴尔分校）

IoT Edge intelligence requires Convolutional Neural Network (CNN) inference to take place in the edge devices itself. ARM big.LITTLE architecture is at the heart of prevalent commercial edge devices. It comprises of single-ISA heterogeneous cores grouped into multiple homogeneous clusters that enable power and performance trade-offs. All cores are expected to be simultaneously employed in inference to attain maximal throughput. However, high communication overhead involved in parallelization of computations from convolution kernels across clusters is detrimental to throughput. We present an al…

2020-01-22cs.DCcs.LGcs.PF

Mixed-TD：具有层特定张量分解的高效神经网络加速器

Mixed-TD: Efficient Neural Network Accelerator with Layer-Specific Tensor Decomposition

Imperial College London（帝国理工学院）

本文提出Mixed-TD框架，通过混合层特定SVD与CPD张量分解，将CNN高效映射至FPGA，实现每DSP吞吐量提升1.73至10.29倍。

2023-06-22cs.ARcs.LG

Language Models for Image Captioning: The Quirks and What Works

Microsoft Research（微软研究院）；University of Washington（华盛顿大学）；University of California at Berkeley（加州大学伯克利分校）

Two recent approaches have achieved state-of-the-art results in image captioning. The first uses a pipelined process where a set of candidate words is generated by a convolutional neural network (CNN) trained on images, and then a maximum entropy (ME) language model is used to arrange these words into a coherent sentence. The second uses the penultimate activation layer of the CNN as input to a recurrent neural network (RNN) that then generates the caption sequence. In this paper, we compare the merits of these different language modeling approaches for the first time by using the same state-o…

2015-10-14cs.AIcs.CLcs.CV

FFCNN: Fast FPGA based Acceleration for Convolution neural network inference

Atos-BULL；Université Paris Est Créteil（巴黎东克雷泰伊大学）

We present a new efficient OpenCL-based Accelerator for large scale Convolutional Neural Networks called Fast Inference on FPGAs for Convolution Neural Network (FFCNN). FFCNN is based on a deeply pipelined OpenCL kernels architecture. As pointed out before, high-level synthesis tools such as the OpenCL framework can easily port codes originally designed for CPUs and GPUs to FPGAs, but it is still difficult to make OpenCL codes run efficiently on FPGAs. This work aims to propose an efficient FPGA implementation of OpenCL High-Performance Computing Applications. To do so, a Data reuse and task m…

2022-08-28cs.AIcs.ARcs.CV

Synergy: A HW/SW Framework for High Throughput CNNs on Embedded Heterogeneous SoC

School of Computing, National University of Singapore（新加坡国立大学计算机学院）

Convolutional Neural Networks (CNN) have been widely deployed in diverse application domains. There has been significant progress in accelerating both their training and inference using high-performance GPUs, FPGAs, and custom ASICs for datacenter-scale environments. The recent proliferation of mobile and IoT devices have necessitated real-time, energy-efficient deep neural network inference on embedded-class, resource-constrained platforms. In this context, we present {\em Synergy}, an automated, hardware-software co-designed, pipelined, high-throughput CNN inference framework on embedded het…

2018-03-28cs.ARcs.DCcs.LG