Výsledky vyhledávání - "Atici, Salih"

Report

Stein Variational Gradient Descent-based Detection For Random Access With Preambles In MTC

Autor: Zhu, Xin, Pan, Hongyi, Atici, Salih, Cetin, Ahmet Enis

Traditional preamble detection algorithms have low accuracy in the grant-based random access scheme in massive machine-type communication (mMTC). We present a novel preamble detection algorithm based on Stein variational gradient descent (SVGD) at th

Externí odkaz: http://arxiv.org/abs/2309.08782

Zobrazit plný text záznamu

Report

A Hybrid Quantum-Classical Approach based on the Hadamard Transform for the Convolutional Layer

Autor: Pan, Hongyi, Zhu, Xin, Atici, Salih, Cetin, Ahmet Enis

In this paper, we propose a novel Hadamard Transform (HT)-based neural network layer for hybrid quantum-classical computing. It implements the regular convolutional layers in the Hadamard transform domain. The idea is based on the HT convolution theo

Externí odkaz: http://arxiv.org/abs/2305.17510

Zobrazit plný text záznamu

Report

Multichannel Orthogonal Transform-Based Perceptron Layers for Efficient ResNets

Autor: Pan, Hongyi, Hamdan, Emadeldeen, Zhu, Xin, Atici, Salih, Cetin, Ahmet Enis

In this paper, we propose a set of transform-based neural network layers as an alternative to the $3\times3$ Conv2D layers in Convolutional Neural Networks (CNNs). The proposed layers can be implemented based on orthogonal transforms such as the Disc

Externí odkaz: http://arxiv.org/abs/2303.06797

Zobrazit plný text záznamu

Report

Input Normalized Stochastic Gradient Descent Training of Deep Neural Networks

Autor: Atici, Salih, Pan, Hongyi, Cetin, Ahmet Enis

In this paper, we propose a novel optimization algorithm for training machine learning models called Input Normalized Stochastic Gradient Descent (INSGD), inspired by the Normalized Least Mean Squares (NLMS) algorithm used in adaptive filtering. When

Externí odkaz: http://arxiv.org/abs/2212.09921

Zobrazit plný text záznamu

Report

DCT Perceptron Layer: A Transform Domain Approach for Convolution Layer

Autor: Pan, Hongyi, Zhu, Xin, Atici, Salih, Cetin, Ahmet Enis

In this paper, we propose a novel Discrete Cosine Transform (DCT)-based neural network layer which we call DCT-perceptron to replace the $3\times3$ Conv2D layers in the Residual neural Network (ResNet). Convolutional filtering operations are performe

Externí odkaz: http://arxiv.org/abs/2211.08577

Zobrazit plný text záznamu

Report

Classification of the Cervical Vertebrae Maturation (CVM) stages Using the Tripod Network

Autor: Atici, Salih, Pan, Hongyi, Elnagar, Mohammed H., Allareddy, Veerasathpurush, Suhaym, Omar, Ansari, Rashid, Cetin, Ahmet Enis

We present a novel deep learning method for fully automated detection and classification of the Cervical Vertebrae Maturation (CVM) stages. The deep convolutional neural network consists of three parallel networks (TriPodNet) independently trained wi

Externí odkaz: http://arxiv.org/abs/2211.08505

Zobrazit plný text záznamu

Report

Multipod Convolutional Network

Autor: Pan, Hongyi, Atici, Salih, Cetin, Ahmet Enis

In this paper, we introduce a convolutional network which we call MultiPodNet consisting of a combination of two or more convolutional networks which process the input image in parallel to achieve the same goal. Output feature maps of parallel convol

Externí odkaz: http://arxiv.org/abs/2210.00689

Zobrazit plný text záznamu