Direct Convolution

Direct convolution, a fundamental operation in convolutional neural networks (CNNs), is being actively researched to improve efficiency and accuracy across diverse applications. Current efforts focus on optimizing convolution through architectural innovations like dilated convolutions, attention mechanisms integrated with convolutions, and novel data layouts for improved hardware performance, as well as exploring alternatives to traditional convolutions, such as using semirings or table lookups. These advancements aim to enhance the speed and accuracy of CNNs for tasks ranging from medical image analysis and object detection to speech processing and large language model efficiency, ultimately impacting various scientific fields and practical applications.

Papers

January 8, 2023

Equivariant and Steerable Neural Networks: A review with special emphasis on the symmetric group
Patrick Krüger, Hanno Gottschalk
Convolutional Neural Network Narrative Review Direct Convolution Equivariant Map Equivariant Layer Special Emphasis Steerable Convolution

January 1, 2023

GoogLe2Net: Going Transverse with Convolutions
Yuanpeng He
Convolutional Neural Network Direct Convolution Convolutional Layer CNN Architecture

December 31, 2022

Identification of lung nodules CT scan using YOLOv5 based on convolution neural network
Haytham Al Ewaidat, Youness El Brag
Neural Network Person Identification Direct Convolution Fire YOLOv5 Attains Nodule Detection Lung Nodule Lung Nodule Localization

December 23, 2022

A Close Look at Spatial Modeling: From Attention to Convolution
Xu Ma, Huan Wang, Can Qin, Kunpeng Li, Xingchen Zhao, Jie Fu, Yun Fu
Vision Transformer Self Attention Human Attention Direct Convolution Convolutional Layer Attention Map Closer Look Spatial Modelling

December 2, 2022

Dunhuang murals contour generation network based on convolution and self-attention fusion
Baokai Liu, Fengjie He, Shiqiang Du, Kaiwu Zhang, Jianhua Wang
Self Attention Direct Convolution Edge Information Edge Map Contour Initialization Three Dimensional Contour Optimisation

December 1, 2022

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics
Saurabh Deshpande, Raúl I. Sosa, Stéphane P. A. Bordas, Jakub Lengiewicz
Deep Neural Network Human Attention Simulation Study Surrogate Model Direct Convolution Solid Mechanic Complex Deformation

November 22, 2022

November 15, 2022

Scalar Invariant Networks with Zero Bias
Chuqin Geng, Xiaojie Xu, Haolin Ye, Xujie Si
Image Classification Direct Convolution Invariant Network Bias Free ReLU Network

November 10, 2022

Demystify Transformers & Convolutions in Modern Image Deep Networks
Xiaowei Hu, Min Shi, Weiyun Wang, Sitong Wu, Linjie Xing, Wenhai Wang, Xizhou Zhu, Lewei Lu, Jie Zhou, Xiaogang Wang, Yu Qiao, Jifeng Dai
Vision Transformer Direct Convolution Receptive Field Traditional Convolution Attention Operator

November 9, 2022

October 31, 2022

Towards Relation-centered Pooling and Convolution for Heterogeneous Graph Learning Networks
Tiehua Zhang, Yuze Liu, Yao Yao, Youhua Xia, Xin Chen, Xiaowei Huang, Jiong Jin
Graph Representation Learning Direct Convolution Heterogeneous Graph Graph Learning Model Heterogeneous Graph Learning

October 25, 2022

A deep learning approach for brain tumor detection using magnetic resonance imaging
Al-Akhir Nayan, Ahamad Nokib Mozumder, Md. Rakibul Haque, Fahim Hossain Sifat, Khan Raqib Mahmud, Abul Kalam Al Azad, Muhammad Golam Kibria
Neural Network Magnetic Resonance Imaging Deep Learning Approach Direct Convolution Convolution Layer Brain Tumor Brain Tumor Detection

October 18, 2022

Analysis of Convolutions, Non-linearity and Depth in Graph Neural Networks using Neural Tangent Kernel
Mahalakshmi Sabanayagam, Pascal Esser, Debarghya Ghoshdastidar
Graph Neural Network Direct Convolution Neural Tangent Kernel Graph Convolution Large Depth Non Linearity Graph Neural Tangent Kernel

October 17, 2022

October 16, 2022

Demystifying CNNs for Images by Matched Filters
Shengxi Li, Xinyi Zhao, Ljubisa Stankovic, Danilo Mandic
Neural Network CNN Model Direct Convolution Convolution Operation

October 15, 2022

Aplicaci\'on de redes neuronales convolucionales profundas al diagn\'ostico asistido de la enfermedad de Alzheimer
Ángel de la Vega Jiménez
Convolutional Neural Network Data Augmentation Alzheimer'S Disease Direct Convolution Brain Image

October 9, 2022

Strong Gravitational Lensing Parameter Estimation with Vision Transformer
Kuan-Wei Huang, Geoff Chih-Fan Chen, Po-Wen Chang, Sheng-Chieh Lin, Chia-Jung Hsu, Vishal Thengane, Joshua Yao-Yu Lin
Vision Transformer Direct Convolution Gravitational Lens Quasar Host Galaxy

Direct Convolution

Papers

Equivariant and Steerable Neural Networks: A review with special emphasis on the symmetric group

GoogLe2Net: Going Transverse with Convolutions

Identification of lung nodules CT scan using YOLOv5 based on convolution neural network

A Close Look at Spatial Modeling: From Attention to Convolution

Dunhuang murals contour generation network based on convolution and self-attention fusion

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics

Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring

Convolutional Neural Generative Coding: Scaling Predictive Coding to Natural Images

Scalar Invariant Networks with Zero Bias

Demystify Transformers & Convolutions in Modern Image Deep Networks

QuadConv: Quadrature-Based Convolutions with Applications to Non-Uniform PDE Data Compression

LiCo-Net: Linearized Convolution Network for Hardware-efficient Keyword Spotting

Towards Relation-centered Pooling and Convolution for Heterogeneous Graph Learning Networks

A deep learning approach for brain tumor detection using magnetic resonance imaging

Analysis of Convolutions, Non-linearity and Depth in Graph Neural Networks using Neural Tangent Kernel

Deformably-Scaled Transposed Convolution

Packed-Ensembles for Efficient Uncertainty Estimation

Demystifying CNNs for Images by Matched Filters

Aplicaci\'on de redes neuronales convolucionales profundas al diagn\'ostico asistido de la enfermedad de Alzheimer

Strong Gravitational Lensing Parameter Estimation with Vision Transformer