"합성곱 신경망"의 두 판 사이의 차이

수학노트
둘러보기로 가기 검색하러 가기
(→‎노트: 새 문단)
 
 
(같은 사용자의 중간 판 5개는 보이지 않습니다)
1번째 줄: 1번째 줄:
 
== 노트 ==
 
== 노트 ==
  
−
* The architecture of a CNN is designed to take advantage of the 2D structure of an input image (or other 2D input such as a speech signal).<ref name="ref_7639">[http://deeplearning.stanford.edu/tutorial/supervised/ConvolutionalNeuralNetwork/ Unsupervised Feature Learning and Deep Learning Tutorial]</ref>
+
===위키데이터===
−
* A CNN consists of a number of convolutional and subsampling layers optionally followed by fully connected layers.<ref name="ref_7639" />
+
* ID :  [https://www.wikidata.org/wiki/Q17084460 Q17084460]
−
* The figure below illustrates a full layer in a CNN consisting of convolutional and subsampling sublayers.<ref name="ref_7639" />
+
===말뭉치===
−
* As the number of filters (output feature map depth) applied to the input increases, so does the number of features the CNN can extract.<ref name="ref_0526">[https://developers.google.com/machine-learning/practica/image-classification/convolutional-neural-networks ML Practicum: Image Classification]</ref>
+
# ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture.<ref name="ref_ed02149b">[https://cs231n.github.io/convolutional-networks/ CS231n Convolutional Neural Networks for Visual Recognition]</ref>
−
* Other CNNs may contain larger or smaller numbers of convolutional modules, and greater or fewer fully connected layers.<ref name="ref_0526" />
+
# A ConvNet arranges its neurons in three dimensions (width, height, depth), as visualized in one of the layers.<ref name="ref_ed02149b" />
−
* Besides, this article presents the fundamentals, tools, and two examples of the use of CNNs for fruit sorting and quality control.<ref name="ref_66bd">[https://www.mdpi.com/2076-3417/10/10/3443 A Review of Convolutional Neural Network Applied to Fruit Image Processing]</ref>
+
# A ConvNet is made up of Layers.<ref name="ref_ed02149b" />
−
* A convolutional neural network, or CNN, is a deep learning neural network designed for processing structured arrays of data such as images.<ref name="ref_d1ae">[https://deepai.org/machine-learning-glossary-and-terms/convolutional-neural-network Convolutional Neural Network]</ref>
+
# We use three main types of layers to build ConvNet architectures: Convolutional Layer, Pooling Layer, and Fully-Connected Layer (exactly as seen in regular Neural Networks).<ref name="ref_ed02149b" />
−
* It is this property that makes convolutional neural networks so powerful for computer vision.<ref name="ref_d1ae" />
+
# The final layer of the CNN architecture uses a classification layer such as softmax to provide the classification output.<ref name="ref_5edec40d">[https://kr.mathworks.com/discovery/convolutional-neural-network-matlab.html Convolutional Neural Network]</ref>
−
* The activation function has the effect of adding non-linearity into the convolutional neural network.<ref name="ref_d1ae" />
+
# This tutorial demonstrates training a simple Convolutional Neural Network (CNN) to classify CIFAR images.<ref name="ref_12404c57">[https://www.tensorflow.org/tutorials/images/cnn Convolutional Neural Network (CNN)]</ref>
−
* By the tenth layer, a convolutional neural network is able to detect more complex shapes such as eyes.<ref name="ref_d1ae" />
+
# As input, a CNN takes tensors of shape (image_height, image_width, color_channels), ignoring the batch size.<ref name="ref_12404c57" />
−
* Intuitively, this is because a convolutional neural network should be able to detect features in an image no matter where they are located.<ref name="ref_d1ae" />
+
# In this example, you will configure our CNN to process inputs of shape (32, 32, 3), which is the format of CIFAR images.<ref name="ref_12404c57" />
−
* Note that the final layer of a convolutional neural network is normally fully connected.<ref name="ref_d1ae" />
+
# Why ReLU is important : ReLU’s purpose is to introduce non-linearity in our ConvNet.<ref name="ref_b3925190">[https://medium.com/@RaghavPrabhu/understanding-of-convolutional-neural-network-cnn-deep-learning-99760835f148 Understanding of Convolutional Neural Network (CNN) — Deep Learning]</ref>
−
* A Convolutional Neural Network (CNN) is a deep learning algorithm that can recognize and classify features in images for computer vision.<ref name="ref_7a43">[https://missinglink.ai/guides/convolutional-neural-networks/convolutional-neural-network-architecture-forging-pathways-future/ Convolutional Neural Network Architecture: Forging Pathways to the Future]</ref>
+
# The pre-processing required in a ConvNet is much lower as compared to other classification algorithms.<ref name="ref_45f9e918">[https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53 A Comprehensive Guide to Convolutional Neural Networks — the ELI5 way]</ref>
−
* The architecture of a CNN is a key factor in determining its performance and efficiency.<ref name="ref_7a43" />
+
# The architecture of a ConvNet is analogous to that of the connectivity pattern of Neurons in the Human Brain and was inspired by the organization of the Visual Cortex.<ref name="ref_45f9e918" />
−
* This 7-layer CNN classified digits, digitized 32×32 pixel greyscale input images.<ref name="ref_7a43" />
+
# A ConvNet is able to successfully capture the Spatial and Temporal dependencies in an image through the application of relevant filters.<ref name="ref_45f9e918" />
−
* GoogleNet (2014) Built with a CNN inspired by LetNet, the GoogleNet network, which is also named Inception V1, was made by a team at Google.<ref name="ref_7a43" />
+
# The role of the ConvNet is to reduce the images into a form which is easier to process, without losing features which are critical for getting a good prediction.<ref name="ref_45f9e918" />
−
* LeNet was one of the very first convolutional neural networks which helped propel the field of Deep Learning.<ref name="ref_e43f">[https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/ An Intuitive Explanation of Convolutional Neural Networks]</ref>
+
# CNN is a class of deep learning networks that has attracted much attention in recent studies.<ref name="ref_84209ded">[https://www.sciencedirect.com/topics/engineering/convolutional-neural-network Convolutional Neural Network - an overview]</ref>
−
* The primary purpose of Convolution in case of a ConvNet is to extract features from the input image.<ref name="ref_e43f" />
+
# At the beginning of the application of CNN, a similar pipeline is adopted as in the supervised learning models, except that the machine learning algorithms are replaced by CNN.<ref name="ref_84209ded" />
−
* It is important to understand that these layers are the basic building blocks of any CNN.<ref name="ref_e43f" />
+
# The pipeline of CNN-based models is illustrated in Fig.<ref name="ref_84209ded" />
−
* Please note however, that these operations can be repeated any number of times in a single ConvNet.<ref name="ref_e43f" />
+
# Illustration of CNN-based model.<ref name="ref_84209ded" />
−
* Visualizing a ConvNet trained on handwritten digits.<ref name="ref_e43f" />
+
# As shown, it is necessary to prepare a large number of training data with corresponding labels for efficient classification using CNN.<ref name="ref_6e8aae44">[https://insightsimaging.springeropen.com/articles/10.1007/s13244-018-0639-9 Convolutional neural networks: an overview and application in radiology]</ref>
−
* In this post, I have tried to explain the main concepts behind Convolutional Neural Networks in simple terms.<ref name="ref_e43f" />
+
# They used a multiview strategy in 3D-CNN, whose inputs were obtained by cropping three 3D patches of a lung nodule in different sizes and then resizing them into the same size.<ref name="ref_6e8aae44" />
−
* Convolutional neural networks apply a series of learnable filters to the input image.<ref name="ref_bab0">[https://codelabs.developers.google.com/codelabs/keras-flowers-convnets Convolutional neural networks, with Keras and TPUs]</ref>
+
# To utilize time series data, the study used triphasic CT images as 2D images with three channels, which corresponds to the RGB color channels in computer vision, for 2D-CNN.<ref name="ref_6e8aae44" />
−
* This is how a simple convolutional neural network looks in Keras: model = tf.keras.<ref name="ref_bab0" />
+
# However, one can also apply CNN to this task as well.<ref name="ref_6e8aae44" />
−
* Illustration: a convolutional neural network transforms "cubes" of data into other "cubes" of data.<ref name="ref_bab0" />
+
# Central to the convolutional neural network is the convolutional layer that gives the network its name.<ref name="ref_94c67b81">[https://machinelearningmastery.com/convolutional-layers-for-deep-learning-neural-networks/ How Do Convolutional Layers Work in Deep Learning Neural Networks?]</ref>
−
* As you explore in-depth, computer vision and deep learning become all about the convolutional neural network (CNN).<ref name="ref_a315">[https://www.thinkautomation.com/eli5/eli5-what-is-a-convolutional-neural-network/ ELI5: what is a convolutional neural network?]</ref>
+
# In the context of a convolutional neural network, a convolution is a linear operation that involves the multiplication of a set of weights with the input, much like a traditional neural network.<ref name="ref_94c67b81" />
−
* Convolutional neural networks were inspired by animal vision.<ref name="ref_a315" />
+
# The name “convolutional neural network” indicates that the network employs a mathematical operation called convolution.<ref name="ref_2da42839">[https://en.wikipedia.org/wiki/Convolutional_neural_network Convolutional neural network]</ref>
−
* Convolutional layers are the layers that give convolutional neural networks the name.<ref name="ref_a315" />
+
# A convolutional neural network consists of an input and an output layer, as well as multiple hidden layers.<ref name="ref_2da42839" />
−
* In other words, pooling layers give flexibility to your convolutional neural network.<ref name="ref_a315" />
+
# The hidden layers of a CNN typically consist of a series of convolutional layers that convolve with a multiplication or other dot product.<ref name="ref_2da42839" />
−
* This is where all the features extracted by the convolutional neural network get combined.<ref name="ref_a315" />
+
# The neocognitron is the first CNN which requires units located at multiple network positions to have shared weights.<ref name="ref_2da42839" />
−
* Convolutional neural networks sit behind a few AI functions.<ref name="ref_a315" />
+
# To start, the CNN receives an input feature map: a three-dimensional matrix where the size of the first two dimensions corresponds to the length and width of the images in pixels.<ref name="ref_ddd0610c">[https://developers.google.com/machine-learning/practica/image-classification/convolutional-neural-networks ML Practicum: Image Classification]</ref>
−
* Convolutional neural networks let computers ‘see’ pictures.<ref name="ref_a315" />
+
# For each filter-tile pair, the CNN performs element-wise multiplication of the filter matrix and the tile matrix, and then sums all the elements of the resulting matrix to get a single value.<ref name="ref_ddd0610c" />
−
* Although much less common, CNNs are also being looked at to help with video analysis.<ref name="ref_a315" />
+
# During training, the CNN "learns" the optimal values for the filter matrices that enable it to extract meaningful features (textures, edges, shapes) from the input feature map.<ref name="ref_ddd0610c" />
−
* This overview has only scratched the surface of convolutional neural networks.<ref name="ref_a315" />
+
# As the number of filters (output feature map depth) applied to the input increases, so does the number of features the CNN can extract.<ref name="ref_ddd0610c" />
−
* Abstract Deep convolutional neural networks have been successfully applied to many image-processing problems in recent works.<ref name="ref_5eff">[https://www.pnas.org/content/115/2/254 A mixed-scale dense convolutional neural network for image analysis]</ref>
+
# These operations are the basic building blocks of every Convolutional Neural Network, so understanding how these work is an important step to developing a sound understanding of ConvNets.<ref name="ref_e4795691">[https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/ An Intuitive Explanation of Convolutional Neural Networks]</ref>
−
* We first introduce notation and discuss the general structure of existing deep convolutional networks.<ref name="ref_5eff" />
+
# The primary purpose of Convolution in case of a ConvNet is to extract features from the input image.<ref name="ref_e4795691" />
−
* Convolutional neural networks (CNNs) model the unknown function f by using several layers that are connected to each other in succession.<ref name="ref_5eff" />
+
# It is important to understand that these layers are the basic building blocks of any CNN.<ref name="ref_e4795691" />
−
* A schematic of a two-layer CNN architecture is shown in Fig.<ref name="ref_5eff" />
+
# Please note however, that these operations can be repeated any number of times in a single ConvNet.<ref name="ref_e4795691" />
−
* such that the CNN performs the task that is required.<ref name="ref_5eff" />
+
# Deep CNN made a noteworthy contribution in several domains like image classification and recognition; therefore, they become widely known standards.<ref name="ref_af995ae6">[https://link.springer.com/article/10.1007/s13748-019-00203-0 Convolutional neural network: a review of models, methodologies and applications to object detection]</ref>
−
* In general, the increased depth of DCNNs compared with shallow CNNs makes training more difficult.<ref name="ref_5eff" />
+
# This section describes various classical and modern architectures of Deep CNN, which are currently utilized as a building block of several segmentation architectures.<ref name="ref_af995ae6" />
−
* In this section, we describe a new convolutional neural network (CNN) based on the SMILES notation of compounds.<ref name="ref_5dae">[https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-018-2523-5 Convolutional neural network based on SMILES representation of compounds for detecting chemical motif]</ref>
+
# This CNN model implements dropout layers with a particular end goal to battle the issue of overfitting to the training data.<ref name="ref_af995ae6" />
−
* An overview of our CNN is shown in Fig.<ref name="ref_5dae" />
+
# The controller is a predefined RNN, where child model is the required CNN for classification of images.<ref name="ref_af995ae6" />
−
* Our CNN transforms the SMILES feature matrix into a low-dimensional feature vector termed the SMILES convolution finger print (SCFP).<ref name="ref_5dae" />
+
# Rather than focus on one pixel at a time, a convolutional net takes in square patches of pixels and passes them through a filter.<ref name="ref_d0932993">[https://wiki.pathmind.com/convolutional-network A Beginner's Guide to Convolutional Neural Networks (CNNs)]</ref>
−
* CNN has multiple layers consisting of two convolutional and pooling layers with a subsequent global pooling layer.<ref name="ref_5dae" />
+
# The next layer in a convolutional network has three names: max pooling, downsampling and subsampling.<ref name="ref_d0932993" />
−
* Figure 2 shows the architecture of our CNN.<ref name="ref_5dae" />
+
# As more and more information is lost, the patterns processed by the convolutional net become more abstract and grow more distant from visual patterns we recognize as humans.<ref name="ref_d0932993" />
−
* Our CNN has several hyperparameters including the window size of filters, the number of filters, and others.<ref name="ref_5dae" />
+
# To teach an algorithm how to recognise objects in images, we use a specific type of Artificial Neural Network: a Convolutional Neural Network (CNN).<ref name="ref_057251a0">[https://www.freecodecamp.org/news/an-intuitive-guide-to-convolutional-neural-networks-260c2de0a050/ An intuitive guide to Convolutional Neural Networks]</ref>
−
* Our CNN can be used not only as a prediction method but also as a method to compute a fingerprint.<ref name="ref_5dae" />
+
# In the case of a Convolutional Neural Network, the output of the convolution will be passed through the activation function.<ref name="ref_057251a0" />
−
* Koushik, J.: Understanding convolutional neural networks.<ref name="ref_6dee">[https://link.springer.com/article/10.1007/s13748-019-00203-0 Convolutional neural network: a review of models, methodologies and applications to object detection]</ref>
+
# After a convolution layer, it is common to add a pooling layer in between CNN layers.<ref name="ref_057251a0" />
−
* Gated spatio and temporal convolutional neural network for activity recognition: towards gated multimodal deep learning.<ref name="ref_6dee" />
+
# Training a CNN works in the same way as a regular neural network, using backpropagration or gradient descent.<ref name="ref_057251a0" />
−
* A review of object detection based on convolutional neural network.<ref name="ref_6dee" />
+
# A convolutional neural network, or CNN, is a deep learning neural network designed for processing structured arrays of data such as images.<ref name="ref_d1aed079">[https://deepai.org/machine-learning-glossary-and-terms/convolutional-neural-network Convolutional Neural Network]</ref>
−
* Xu, H., Han, Z., Feng, S., Zhou, H., Fang, Y.: Foreign object debris material recognition based on convolutional neural networks.<ref name="ref_6dee" />
+
# The architecture of a convolutional neural network is a multi-layered feed-forward neural network, made by stacking many hidden layers on top of each other in sequence.<ref name="ref_d1aed079" />
−
* Rethinking Model Scaling for Convolutional Neural Networks.<ref name="ref_6dee" />
+
# A simple convolutional neural network that aids understanding of the core design principles is the early convolutional neural network LeNet-5, published by Yann LeCun in 1998.<ref name="ref_d1aed079" />
−
* CNNs are not limited to image recognition, however.<ref name="ref_59a8">[https://wiki.pathmind.com/convolutional-network A Beginner's Guide to Convolutional Neural Networks (CNNs)]</ref>
+
# Typically, the first layer of a convolutional neural network contains a vertical line detector, a horizontal line detector, and various diagonal, curve and corner detectors.<ref name="ref_d1aed079" />
−
* So instead of thinking of images as two-dimensional areas, in convolutional nets they are treated as four-dimensional volumes.<ref name="ref_59a8" />
+
# The neurons within a CNN are split into a three-dimensional structure, with each set of neurons analyzing a small region or feature of the image.<ref name="ref_53b22d0c">[https://missinglink.ai/guides/convolutional-neural-networks/convolutional-neural-network-architecture-forging-pathways-future/ Convolutional Neural Network Architecture: Forging Pathways to the Future]</ref>
−
* (Note that convolutional nets analyze images differently than RBMs.<ref name="ref_59a8" />
+
# A Convolutional Neural Network (CNN) is a deep learning algorithm that can recognize and classify features in images for computer vision.<ref name="ref_53b22d0c" />
−
* So convolutional networks perform a sort of search.<ref name="ref_59a8" />
+
# The architecture of a CNN is a key factor in determining its performance and efficiency.<ref name="ref_53b22d0c" />
−
* The first thing to know about convolutional networks is that they don’t perceive images like humans do.<ref name="ref_59a8" />
+
# The ImageNet project has more than 14 million images specifically designed for training CNN in object detection, one million of which also provide bounding boxes for the use of networks such as YOLO.<ref name="ref_53b22d0c" />
−
* Convolutional networks are designed to reduce the dimensionality of images in a variety of ways.<ref name="ref_59a8" />
+
# Our study proposes a prediction method using one-dimensional convolutional neural network (1-D CNN) that contains all aforementioned processes together.<ref name="ref_79bd5325">[https://www.nature.com/articles/s41598-020-63566-8 Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features]</ref>
−
* The next layer in a convolutional network has three names: max pooling, downsampling and subsampling.<ref name="ref_59a8" />
+
# With the advent of deep learning, 1-D CNN has become favorable for extracting features from time series signals, and therefore for detection, prediction, and classification10.<ref name="ref_79bd5325" />
−
* Prior to CNNs, manual, time-consuming feature extraction methods were used to identify objects in images.<ref name="ref_e497">[https://www.ibm.com/cloud/learn/convolutional-neural-networks What are Convolutional Neural Networks?]</ref>
+
# Therefore, CNN possesses the capacity to extract features from the 1-D time series data of raw ECG signals and use them to monitor mental stress and detect myocardial infractions (MI)11.<ref name="ref_79bd5325" />
−
* With each layer, the CNN increases in its complexity, identifying greater portions of the image.<ref name="ref_e497" />
+
# We propose a prediction method for VT and VF based on a 1-D CNN trained using HRV signals.<ref name="ref_79bd5325" />
−
* The convolutional layer is the core building block of a CNN, and it is where the majority of computation occurs.<ref name="ref_e497" />
+
# When you input an image into a ConvNet, each of its layers generates several activation maps.<ref name="ref_0fb63c57">[https://bdtechtalks.com/2020/01/06/convolutional-neural-networks-cnn-convnets/ What are convolutional neural networks (CNN)?]</ref>
−
* While a lot of information is lost in the pooling layer, it also has a number of benefits to the CNN.<ref name="ref_e497" />
+
# The first (or bottom) layer of the CNN usually detects basic features such as horizontal, vertical, and diagonal edges.<ref name="ref_0fb63c57" />
−
* Both the CNN technique and the previous pre-computed brain response atlas approach15 require a large training dataset.<ref name="ref_620c">[https://www.nature.com/articles/s41598-019-53551-1 Convolutional neural network for efficient estimation of regional brain strains]</ref>
+
# The operation of multiplying pixel values by weights and summing them is called “convolution” (hence the name convolutional neural network).<ref name="ref_0fb63c57" />
−
* This indicated some impressive generalizability and robustness of the CNN technique.<ref name="ref_620c" />
+
# A CNN is usually composed of several convolution layers, but it also contains other components.<ref name="ref_0fb63c57" />
−
* Further, the data-driven CNN technique does not address any physics behind brain biomechanical responses.<ref name="ref_620c" />
+
# In this section, we describe a new convolutional neural network (CNN) based on the SMILES notation of compounds.<ref name="ref_5daea7c8">[https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-018-2523-5 Convolutional neural network based on SMILES representation of compounds for detecting chemical motif]</ref>
−
* Nonetheless, the CNN can be easily re-trained to accommodate another model or a future, upgraded WHIM.<ref name="ref_620c" />
+
# An overview of our CNN is shown in Fig.<ref name="ref_5daea7c8" />
−
* In recent years, CNNs have become pivotal to many computer vision applications.<ref name="ref_cf6a">[https://bdtechtalks.com/2020/01/06/convolutional-neural-networks-cnn-convnets/ What are convolutional neural networks (CNN)?]</ref>
+
# Our CNN transforms the SMILES feature matrix into a low-dimensional feature vector termed the SMILES convolution finger print (SCFP).<ref name="ref_5daea7c8" />
−
* The early version of CNNs, called LeNet (after LeCun), could recognize handwritten digits.<ref name="ref_cf6a" />
+
# CNN has multiple layers consisting of two convolutional and pooling layers with a subsequent global pooling layer.<ref name="ref_5daea7c8" />
−
* CNNs needed a lot of data and compute resources to work efficiently for large images.<ref name="ref_cf6a" />
+
# But what is a convolutional neural network and why has it suddenly become so popular?<ref name="ref_68a5eac1">[https://www.analyticsvidhya.com/blog/2018/12/guide-convolutional-neural-network-cnn/ Tutorial On Convolutional Neural Networks]</ref>
−
* Convolutional neural networks are composed of multiple layers of artificial neurons.<ref name="ref_cf6a" />
+
# In addition to exploring how a convolutional neural network (ConvNet) works, we’ll also look at different architectures of a ConvNet and how we can build an object detection model using YOLO.<ref name="ref_68a5eac1" />
−
* When you input an image into a ConvNet, each of its layers generates several activation maps.<ref name="ref_cf6a" />
+
# There are a number of hyperparameters that we can tweak while building a convolutional network.<ref name="ref_68a5eac1" />
−
* The first (or bottom) layer of the CNN usually detects basic features such as horizontal, vertical, and diagonal edges.<ref name="ref_cf6a" />
+
# Let’s look at how a convolution neural network with convolutional and pooling layer works.<ref name="ref_68a5eac1" />
−
* The operation of multiplying pixel values by weights and summing them is called “convolution” (hence the name convolutional neural network).<ref name="ref_cf6a" />
+
# A Convolutional Neural Network is a class of artificial neural network that uses convolutional layers to filter inputs for useful information.<ref name="ref_cfc25bee">[https://developer.nvidia.com/discover/convolutional-neural-network Convolutional Neural Network (CNN)]</ref>
−
* A CNN is usually composed of several convolution layers, but it also contains other components.<ref name="ref_cf6a" />
+
# The CNN would filter information about the shape of an object when confronted with a general object recognition task but would extract the color of the bird when faced with a bird recognition task.<ref name="ref_cfc25bee" />
−
* One of the great challenges of developing CNNs is adjusting the weights of the individual neurons to extract the right features from images.<ref name="ref_cf6a" />
+
# A convolutional network is different than a regular neural network in that the neurons in its layers are arranged in three dimensions (width, height, and depth dimensions).<ref name="ref_cfc25bee" />
−
* In the beginning, the CNN starts off with random weights.<ref name="ref_cf6a" />
+
# This allows the CNN to transform an input volume in three dimensions to an output volume.<ref name="ref_cfc25bee" />
−
* The ConvNet processes each image with its random values and then compares its output with the image’s correct label.<ref name="ref_cf6a" />
+
# Each kernel can output a feature map and all the feature maps are concatenated together, this is also known as a convolutional layer and it is the core component in a CNN.<ref name="ref_c201d43d">[https://www.intechopen.com/books/advances-and-applications-in-deep-learning/advances-in-convolutional-neural-networks Advances in Convolutional Neural Networks]</ref>
−
* The ConvNet goes through several epochs during training, adjusting its weights in small amounts.<ref name="ref_cf6a" />
+
# A typical CNN is usually structured as a series of layers, including multiple convolutional layers and a few of fully connected layers.<ref name="ref_c201d43d" />
−
* As the CNN improves, the adjustments it makes to the weights become smaller and smaller.<ref name="ref_cf6a" />
+
# Therefore, the process of training a CNN model is transformed into an optimization problem, which normally seeks to minimize the value of the loss function over the training data.<ref name="ref_c201d43d" />
−
* After training the CNN, the developers use a test dataset to verify its accuracy.<ref name="ref_cf6a" />
+
# Specifically, a gradient-descent based algorithm is usually adopted to iteratively optimize the parameters in a CNN.<ref name="ref_c201d43d" />
−
* Each image is run through the ConvNet, and the output is compared to the actual label of the image.<ref name="ref_cf6a" />
+
# A schematic of a two-layer CNN architecture is shown in Fig.<ref name="ref_0f15c6ee">[https://www.pnas.org/content/115/2/254 A mixed-scale dense convolutional neural network for image analysis]</ref>
−
* The success of convolutional neural networks is largely due to the availability of huge image datasets developed in the past decade.<ref name="ref_cf6a" />
+
# such that the CNN performs the task that is required.<ref name="ref_0f15c6ee" />
−
* You don’t, however, need to train every convolutional neural network on millions of images.<ref name="ref_cf6a" />
+
# The main aim of this work is to investigate the applicability of CNN in impact detection and identification of complex composite structures such as aircraft stiffened panel.<ref name="ref_ea8ff06a">[https://www.mdpi.com/1424-8220/19/22/4933/htm A Convolutional Neural Network for Impact Detection and Characterization of Complex Composite Structures]</ref>
−
* A well-trained ConvNet will tell you that it’s the image of a soldier, a child and the American flag.<ref name="ref_cf6a" />
+
# Thus, using a CNN with raw data as input will be more advantageous than traditional extraction methods.<ref name="ref_ea8ff06a" />
−
* These limits become more evident in practical applications of convolutional neural networks.<ref name="ref_cf6a" />
+
# It is used to normalise the output vector of the CNN, which is of length equal to the number of classes, say, to a vector of length, whose values sum to 1.<ref name="ref_ea8ff06a" />
−
* For instance, CNNs are now widely used to moderate content on social media networks.<ref name="ref_cf6a" />
+
# The nodes at the output of the layer, will, thus, contain the probabilities of the input to the CNN belonging to all classes.<ref name="ref_ea8ff06a" />
−
* Another problem with convolutional neural networks is their inability to understand the relations between different objects.<ref name="ref_cf6a" />
+
# In this post I am going to explain what they are and how you can create a convolutional neural network in Keras with Python.<ref name="ref_9845ce6c">[https://anderfernandez.com/en/blog/how-to-create-convolutional-neural-network-keras/ How to build a convolutional neural network in Keras]</ref>
−
* still no convolutional neural network that can solve Bongard problems with so few training examples.<ref name="ref_cf6a" />
+
# We (along with many CNN implementations) are technically actually using cross-correlation instead of convolution here, but they do almost the same thing.<ref name="ref_fb84e2b0">[https://victorzhou.com/blog/intro-to-cnns-part-1/ CNNs, Part 1: An Introduction to Convolutional Neural Networks]</ref>
−
* In one study conducted in 2016, AI researchers trained a CNN on 20,000 Bongard samples and tested it on 10,000 more.<ref name="ref_cf6a" />
+
# A CNN trained on MNIST might look for the digit 1, for example, by using an edge-detection filter and checking for two prominent vertical edges near the center of the image.<ref name="ref_fb84e2b0" />
−
* Today, CNNs are used in many computer vision applications such as facial recognition, image search and editing, augmented reality, and more.<ref name="ref_cf6a" />
+
# For our MNIST CNN, we’ll use a small conv layer with 8 filters as the initial layer in our network.<ref name="ref_fb84e2b0" />
−
* Convolutional neural networks apply neural networks on images.<ref name="ref_b9b1">[https://anderfernandez.com/en/blog/how-to-create-convolutional-neural-network-keras/ How to build a convolutional neural network in Keras]</ref>
+
# For our MNIST CNN, we’ll place a Max Pooling layer with a pool size of 2 right after our initial conv layer.<ref name="ref_fb84e2b0" />
−
* In this post I am going to explain what they are and how you can create a convolutional neural network in Keras with Python.<ref name="ref_b9b1" />
+
# We will conclude the chapter with a full working example of LeNet, the first convolutional network successfully deployed, long before the rise of modern deep learning.<ref name="ref_4b03d8ed">[https://www.d2l.ai/chapter_convolutional-neural-networks/index.html 6. Convolutional Neural Networks — Dive into Deep Learning 0.15.1 documentation]</ref>
−
* Convolutional neural networks basically take an image as input and apply different transformations that condense all the information.<ref name="ref_b9b1" />
+
# To solve the abovementioned two problems, this paper proposes a fault-diagnosis algorithm based on “end-to-end” one-dimensional convolutional neural network.<ref name="ref_da23505c">[https://www.hindawi.com/journals/sv/2020/1971945/ Multitask Convolutional Neural Network for Rolling Element Bearing Fault Identification]</ref>
−
* A typical CNN is usually structured as a series of layers, including multiple convolutional layers and a few of fully connected layers.<ref name="ref_125c">[https://www.intechopen.com/online-first/advances-in-convolutional-neural-networks Advances in Convolutional Neural Networks]</ref>
+
# The model does not rely on artificial feature extraction and expert knowledge and maximizes the use of CNN for feature extraction.<ref name="ref_da23505c" />
−
* Specifically, a gradient-descent based algorithm is usually adopted to iteratively optimize the parameters in a CNN.<ref name="ref_125c" />
+
# Section 2 gives a brief introduction of CNN.<ref name="ref_da23505c" />
−
* Figure 1 shows the high-level abstraction of CNNs in this survey.<ref name="ref_125c" />
+
# The CNN is a special structure of the feed-forward neural network, generally including a convolution layer, activation layer, pooling layer, and fully connected layer.<ref name="ref_da23505c" />
−
* Then four methods are summarized for constructing convolutional layers in CNNs in Section 3.<ref name="ref_125c" />
 
−
* In Section 4, we group the current CNN architectures into three types: encoder, encoder-decoder and GANs.<ref name="ref_125c" />
 
−
* In Section 6, we give the advanced applications based on the three types of CNN structures.<ref name="ref_125c" />
 
−
* However, it is difficult to know what size of kernels we should use in a CNN.<ref name="ref_125c" />
 
−
* Z and F decoder represents a decoder CNN to reconstruct the input sample with Z .<ref name="ref_125c" />
 
−
* Recall that the shortcut connection is often adopted to address the problems in deep CNNs.<ref name="ref_125c" />
 
−
* Therefore, as shown in Figure 1, loss functions play a significant role in constructing CNNs.<ref name="ref_125c" />
 
−
* More Specifically, if Q denotes the distribution on data, and P represents the distribution which is learned by a CNN model.<ref name="ref_125c" />
 
−
* In this section, we summarize the typical advances that CNNs has achieved based on the three types of CNN structures.<ref name="ref_125c" />
 
−
* Firstly, a pre-trained CNN encoder is used to extract some high-level features from an input image.<ref name="ref_125c" />
 
−
* 6.2.4 Speech processing Note that speech signals exhibit spectral variations and correlations, CNNs are very suitable to reduce them.<ref name="ref_125c" />
 
−
* Therefore, CNNs can also be utilized for the task of speech processing, such as speech recognition.<ref name="ref_125c" />
 
−
* Because these two aspects are the core parts when applying CNNs into various types of tasks.<ref name="ref_125c" />
 
−
* Convolutional networks rely on 3D architecture - height, width, and depth - to scale for image recognition.<ref name="ref_316b">[https://www.edx.org/learn/convolutional-neural-network Learn Convolutional Neural Network with Online Courses]</ref>
 
−
* CNN Courses and Certifications Deep neural networks are critical to working with images in the era of visual data science.<ref name="ref_316b" />
 
−
* Building your knowledge of CNN is vital to understanding image data.<ref name="ref_316b" />
 
−
* The building blocks for true robot vision lie in convolutional neural networks where image data can fill in gaps for operation.<ref name="ref_316b" />
 
−
* The specificity of a CNN lies in its filtering layers, which include at least one convolution layer.<ref name="ref_8bf2">[https://www.doc.ic.ac.uk/~jce317/introduction-cnns.html Convolutional Neural Networks]</ref>
 
−
* The first major success of convolutional neural networks was AlexNet, developed by Alex Krizhevsky, in 2012 at the University of Toronto.<ref name="ref_8bf2" />
 
−
* This current state of the art came about from ResNet, a CNN architecture from Microsoft Research.<ref name="ref_8bf2" />
 
−
* The problem motivating ResNet was that adding extra layers to a CNN did not necessarily improve performance.<ref name="ref_8bf2" />
 
−
* D. Wei, B. Zhou, A. Torrabla and W. Freeman, Understanding intra-class knowledge inside cnn, arXiv preprint, arXiv: 1507.02379.<ref name="ref_0379">[https://www.aimsciences.org/article/doi/10.3934/mfc.2018008 How convolutional neural networks see the world --- A survey of convolutional neural network visualization methods]</ref>
 
−
* Visualizing and understanding convolutional networks, in Proceedings of the European Conference on Computer Vision, 2014, 818-833.<ref name="ref_0379" />
 
−
* This paper presents a novel method for identifying coal and rock based on a deep convolutional neural network (CNN).<ref name="ref_6c40">[https://www.hindawi.com/journals/mpe/2020/2616510/ A Deep Convolutional Neural Network Model for Intelligent Discrimination between Coal and Rocks in Coal Mining Face]</ref>
 
−
* Section 2 summarizes the basic theory of CNN.<ref name="ref_6c40" />
 
−
* There are two stages in identifying coal and rock using convolutional neural networks: feature learning and classification.<ref name="ref_6c40" />
 
−
* Figure 2 shows the training process of a convolutional neural network.<ref name="ref_6c40" />
 
−
* In the application of CNN, the problem of overfitting often occurs.<ref name="ref_6c40" />
 
−
* The training of convolutional neural network is the key to achieve the discrimination between coal and rocks.<ref name="ref_6c40" />
 
−
* The raw data (without data augmentation) and augmented data are used to train the CNN with NET, respectively.<ref name="ref_6c40" />
 
−
* This paper presents a method for identifying coal and rock based on a deep convolutional neural network.<ref name="ref_6c40" />
 
−
* Some experiments are provided and the comparisons with other classical convolutional neural networks are conducted.<ref name="ref_6c40" />
 
−
* The convolutional neural network (CNN) was first proposed in 1960s.<ref name="ref_ef11">[https://www.spiedigitallibrary.org/journals/optical-engineering/volume-58/issue-04/040901/Development-of-convolutional-neural-network-and-its-application-in-image/10.1117/1.OE.58.4.040901.full Development of convolutional neural network and its application in image classification: a survey]</ref>
 
−
* 1 was later introduced into the research work of CNNs.<ref name="ref_ef11" />
 
−
* The CNN model AlexNet presented by Krizhevsky et al.<ref name="ref_ef11" />
 
−
* First, this paper introduces the history of CNN and then analyzes the development of CNN architecture in image classification.<ref name="ref_ef11" />
 
−
* 1 , CNN architecture is generally composed of convolution layers, subsampling (pooling) layers, and fully connected layers.<ref name="ref_ef11" />
 
−
* The combination of convolution layer, pooling layer, and fully connected layer is still the basic components of modern deep CNN.<ref name="ref_ef11" />
 
−
* 5 AlexNet is a milestone in the development of deep CNN, which has caused a new wave of neural network research.<ref name="ref_ef11" />
 
−
* After AlexNet achieved excellent results in the ImageNet image classification competition, researchers began to study the CNN more deeply.<ref name="ref_ef11" />
 
−
* However, there is no clear theoretical explanation for why a CNN model can perform well.<ref name="ref_ef11" />
 
−
* Zeiler and Fergus 31 proposed a visualization technique to understand CNNs and proposed ZFNet.<ref name="ref_ef11" />
 
−
* 31 is to explain to a certain extent why CNNs are effective and how to improve network performance.<ref name="ref_ef11" />
 
−
* Using small convolution kernels, VGG can make the CNN reach a depth of 19 layers.<ref name="ref_ef11" />
 
−
* It has made a vital contribution to the development of CNNs.<ref name="ref_ef11" />
 
−
* In addition to network initialization, the innovation of optimization method has also promoted the development of CNN.<ref name="ref_ef11" />
 
−
* Experimental results are obtained through five similar CNN predictions.<ref name="ref_ef11" />
 
−
* In the following, the development trend of CNNs in image classification is prospected through several aspects.<ref name="ref_ef11" />
 
−
* A CNN has a large number of parameters, so the experiment of CNN often fails to achieve the effect of network in corresponding papers.<ref name="ref_ef11" />
 
−
* At present, the parameter setting in training CNN is mostly based on experience and practice.<ref name="ref_ef11" />
 
−
* Visualizing and understanding convolutional networks ,” in Eur.<ref name="ref_ef11" />
 
−
* The final layer of the CNN architecture uses a classification layer such as softmax to provide the classification output.<ref name="ref_5ede">[https://kr.mathworks.com/discovery/convolutional-neural-network-matlab.html Convolutional Neural Network]</ref>
 
−
* This tutorial demonstrates training a simple Convolutional Neural Network (CNN) to classify CIFAR images.<ref name="ref_1240">[https://www.tensorflow.org/tutorials/images/cnn Convolutional Neural Network (CNN)]</ref>
 
−
* As input, a CNN takes tensors of shape (image_height, image_width, color_channels), ignoring the batch size.<ref name="ref_1240" />
 
−
* In this example, you will configure our CNN to process inputs of shape (32, 32, 3), which is the format of CIFAR images.<ref name="ref_1240" />
 
−
* Why ReLU is important : ReLU’s purpose is to introduce non-linearity in our ConvNet.<ref name="ref_b392">[https://medium.com/@RaghavPrabhu/understanding-of-convolutional-neural-network-cnn-deep-learning-99760835f148 Understanding of Convolutional Neural Network (CNN) — Deep Learning]</ref>
 
−
* A ConvNet arranges its neurons in three dimensions (width, height, depth), as visualized in one of the layers.<ref name="ref_be8a">[https://cs231n.github.io/convolutional-networks/ CS231n Convolutional Neural Networks for Visual Recognition]</ref>
 
−
* A ConvNet is made up of Layers.<ref name="ref_be8a" />
 
−
* Together, this adds up to 290400 * 364 = 105,705,600 parameters on the first layer of the ConvNet alone.<ref name="ref_be8a" />
 
−
* Lastly, what if we wanted to efficiently apply the original ConvNet over the image but at a stride smaller than 32 pixels?<ref name="ref_be8a" />
 
−
* You should rarely ever have to train a ConvNet from scratch or design one from scratch.<ref name="ref_be8a" />
 
−
* Until now we’ve omitted mentions of common hyperparameters used in each of the layers in a ConvNet.<ref name="ref_be8a" />
 
−
* The first successful applications of Convolutional Networks were developed by Yann LeCun in 1990’s.<ref name="ref_be8a" />
 
−
* CNN is a class of deep learning networks that has attracted much attention in recent studies.<ref name="ref_8420">[https://www.sciencedirect.com/topics/engineering/convolutional-neural-network Convolutional Neural Network - an overview]</ref>
 
−
* The input matrix of the CNN model is an L× L×(2×D+C) matrix, where C represents the number of 2-D features.<ref name="ref_8420" />
 
−
* The CNN model is composed of several residual blocks.<ref name="ref_8420" />
 
−
* Finally, a three-layer CNN is employed to predict the final contact map from matrix Mat.<ref name="ref_8420" />
 
−
* As shown, it is necessary to prepare a large number of training data with corresponding labels for efficient classification using CNN.<ref name="ref_6e8a">[https://insightsimaging.springeropen.com/articles/10.1007/s13244-018-0639-9 Convolutional neural networks: an overview and application in radiology]</ref>
 
−
* However, one can also apply CNN to this task as well.<ref name="ref_6e8a" />
 
−
* One way to perform segmentation is to use a CNN classifier for calculating the probability of an organ or anatomical structure.<ref name="ref_6e8a" />
 
−
* a Denoising system with CNN in deployment phase.<ref name="ref_6e8a" />
 
−
* By using this representation, 2D-CNN could be used to classify the reports as pulmonary embolism or not.<ref name="ref_6e8a" />
 
−
* Their results showed that the performance of the CNN model was equivalent to or beyond that of the traditional model.<ref name="ref_6e8a" />
 
−
* The pre-processing required in a ConvNet is much lower as compared to other classification algorithms.<ref name="ref_45f9">[https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53 A Comprehensive Guide to Convolutional Neural Networks — the ELI5 way]</ref>
 
−
* A ConvNet is able to successfully capture the Spatial and Temporal dependencies in an image through the application of relevant filters.<ref name="ref_45f9" />
 
−
* ConvNets need not be limited to only one Convolutional Layer.<ref name="ref_45f9" />
 
−
* CNNs are regularized versions of multilayer perceptrons.<ref name="ref_0883">[https://en.wikipedia.org/wiki/Convolutional_neural_network Convolutional neural network]</ref>
 
−
* CNNs use relatively little pre-processing compared to other image classification algorithms.<ref name="ref_0883" />
 
−
* The name “convolutional neural network” indicates that the network employs a mathematical operation called convolution.<ref name="ref_0883" />
 
−
* A convolutional neural network consists of an input and an output layer, as well as multiple hidden layers.<ref name="ref_0883" />
 
−
* The hidden layers of a CNN typically consist of a series of convolutional layers that convolve with a multiplication or other dot product.<ref name="ref_0883" />
 
−
* Convolutional networks may include local or global pooling layers to streamline the underlying computation.<ref name="ref_0883" />
 
−
* A distinguishing feature of CNNs is that many neurons can share the same filter.<ref name="ref_0883" />
 
−
* The neocognitron introduced the two basic types of layers in CNNs: convolutional layers, and downsampling layers.<ref name="ref_0883" />
 
−
* The neocognitron is the first CNN which requires units located at multiple network positions to have shared weights.<ref name="ref_0883" />
 
−
* The first GPU-implementation of a CNN was described in 2006 by K. Chellapilla et al.<ref name="ref_0883" />
 
−
* Subsequently, a similar GPU-based CNN by Alex Krizhevsky et al.<ref name="ref_0883" />
 
−
* The layers of a CNN have neurons arranged in 3 dimensions: width, height and depth.<ref name="ref_0883" />
 
−
* Distinct types of layers, both locally and completely connected, are stacked to form a CNN architecture.<ref name="ref_0883" />
 
−
* In CNNs, each filter is replicated across the entire visual field.<ref name="ref_0883" />
 
−
* Together, these properties allow CNNs to achieve better generalization on vision problems.<ref name="ref_0883" />
 
−
* Another important concept of CNNs is pooling, which is a form of non-linear down-sampling.<ref name="ref_0883" />
 
−
* This is the idea behind the use of pooling in convolutional neural networks.<ref name="ref_0883" />
 
−
* CNNs use more hyperparameters than a standard multilayer perceptron (MLP).<ref name="ref_0883" />
 
−
* For convolutional networks, the filter size also affects the number of parameters.<ref name="ref_0883" />
 
−
* CNNs are often used in image recognition systems.<ref name="ref_0883" />
 
−
* Compared to image data domains, there is relatively little work on applying CNNs to video classification.<ref name="ref_0883" />
 
−
* However, some extensions of CNNs into the video domain have been explored.<ref name="ref_0883" />
 
−
* CNNs have also been explored for natural language processing.<ref name="ref_0883" />
 
−
* CNNs have been used in drug discovery.<ref name="ref_0883" />
 
−
* CNNs have been used in the game of checkers.<ref name="ref_0883" />
 
−
* CNNs have been used in computer Go.<ref name="ref_0883" />
 
−
* Convolutional neural networks usually require a large amount of training data in order to avoid overfitting.<ref name="ref_0883" />
 
−
* Therefore, they exploit the 2D structure of images, like CNNs do, and make use of pre-training like deep belief networks.<ref name="ref_0883" />
 
−
* Central to the convolutional neural network is the convolutional layer that gives the network its name.<ref name="ref_94c6">[https://machinelearningmastery.com/convolutional-layers-for-deep-learning-neural-networks/ How Do Convolutional Layers Work in Deep Learning Neural Networks?]</ref>
 
−
* Technically, the convolution as described in the use of convolutional neural networks is actually a “cross-correlation”.<ref name="ref_94c6" />
 
 
===소스===
 
===소스===
 
  <references />
 
  <references />
 +
 +
==메타데이터==
 +
===위키데이터===
 +
* ID :  [https://www.wikidata.org/wiki/Q17084460 Q17084460]
 +
===Spacy 패턴 목록===
 +
* [{'LOWER': 'convolutional'}, {'LOWER': 'neural'}, {'LEMMA': 'network'}]
 +
* [{'LEMMA': 'CNN'}]
 +
* [{'LEMMA': 'ConvNet'}]
 +
* [{'LOWER': 'convolutional'}, {'LEMMA': 'Network'}]
 +
* [{'LOWER': 'convolution'}, {'LOWER': 'neural'}, {'LEMMA': 'Network'}]
 +
* [{'LOWER': 'convolutional'}, {'LEMMA': 'net'}]

2021년 2월 17일 (수) 01:58 기준 최신판

노트

위키데이터

말뭉치

  1. ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture.[1]
  2. A ConvNet arranges its neurons in three dimensions (width, height, depth), as visualized in one of the layers.[1]
  3. A ConvNet is made up of Layers.[1]
  4. We use three main types of layers to build ConvNet architectures: Convolutional Layer, Pooling Layer, and Fully-Connected Layer (exactly as seen in regular Neural Networks).[1]
  5. The final layer of the CNN architecture uses a classification layer such as softmax to provide the classification output.[2]
  6. This tutorial demonstrates training a simple Convolutional Neural Network (CNN) to classify CIFAR images.[3]
  7. As input, a CNN takes tensors of shape (image_height, image_width, color_channels), ignoring the batch size.[3]
  8. In this example, you will configure our CNN to process inputs of shape (32, 32, 3), which is the format of CIFAR images.[3]
  9. Why ReLU is important : ReLU’s purpose is to introduce non-linearity in our ConvNet.[4]
  10. The pre-processing required in a ConvNet is much lower as compared to other classification algorithms.[5]
  11. The architecture of a ConvNet is analogous to that of the connectivity pattern of Neurons in the Human Brain and was inspired by the organization of the Visual Cortex.[5]
  12. A ConvNet is able to successfully capture the Spatial and Temporal dependencies in an image through the application of relevant filters.[5]
  13. The role of the ConvNet is to reduce the images into a form which is easier to process, without losing features which are critical for getting a good prediction.[5]
  14. CNN is a class of deep learning networks that has attracted much attention in recent studies.[6]
  15. At the beginning of the application of CNN, a similar pipeline is adopted as in the supervised learning models, except that the machine learning algorithms are replaced by CNN.[6]
  16. The pipeline of CNN-based models is illustrated in Fig.[6]
  17. Illustration of CNN-based model.[6]
  18. As shown, it is necessary to prepare a large number of training data with corresponding labels for efficient classification using CNN.[7]
  19. They used a multiview strategy in 3D-CNN, whose inputs were obtained by cropping three 3D patches of a lung nodule in different sizes and then resizing them into the same size.[7]
  20. To utilize time series data, the study used triphasic CT images as 2D images with three channels, which corresponds to the RGB color channels in computer vision, for 2D-CNN.[7]
  21. However, one can also apply CNN to this task as well.[7]
  22. Central to the convolutional neural network is the convolutional layer that gives the network its name.[8]
  23. In the context of a convolutional neural network, a convolution is a linear operation that involves the multiplication of a set of weights with the input, much like a traditional neural network.[8]
  24. The name “convolutional neural network” indicates that the network employs a mathematical operation called convolution.[9]
  25. A convolutional neural network consists of an input and an output layer, as well as multiple hidden layers.[9]
  26. The hidden layers of a CNN typically consist of a series of convolutional layers that convolve with a multiplication or other dot product.[9]
  27. The neocognitron is the first CNN which requires units located at multiple network positions to have shared weights.[9]
  28. To start, the CNN receives an input feature map: a three-dimensional matrix where the size of the first two dimensions corresponds to the length and width of the images in pixels.[10]
  29. For each filter-tile pair, the CNN performs element-wise multiplication of the filter matrix and the tile matrix, and then sums all the elements of the resulting matrix to get a single value.[10]
  30. During training, the CNN "learns" the optimal values for the filter matrices that enable it to extract meaningful features (textures, edges, shapes) from the input feature map.[10]
  31. As the number of filters (output feature map depth) applied to the input increases, so does the number of features the CNN can extract.[10]
  32. These operations are the basic building blocks of every Convolutional Neural Network, so understanding how these work is an important step to developing a sound understanding of ConvNets.[11]
  33. The primary purpose of Convolution in case of a ConvNet is to extract features from the input image.[11]
  34. It is important to understand that these layers are the basic building blocks of any CNN.[11]
  35. Please note however, that these operations can be repeated any number of times in a single ConvNet.[11]
  36. Deep CNN made a noteworthy contribution in several domains like image classification and recognition; therefore, they become widely known standards.[12]
  37. This section describes various classical and modern architectures of Deep CNN, which are currently utilized as a building block of several segmentation architectures.[12]
  38. This CNN model implements dropout layers with a particular end goal to battle the issue of overfitting to the training data.[12]
  39. The controller is a predefined RNN, where child model is the required CNN for classification of images.[12]
  40. Rather than focus on one pixel at a time, a convolutional net takes in square patches of pixels and passes them through a filter.[13]
  41. The next layer in a convolutional network has three names: max pooling, downsampling and subsampling.[13]
  42. As more and more information is lost, the patterns processed by the convolutional net become more abstract and grow more distant from visual patterns we recognize as humans.[13]
  43. To teach an algorithm how to recognise objects in images, we use a specific type of Artificial Neural Network: a Convolutional Neural Network (CNN).[14]
  44. In the case of a Convolutional Neural Network, the output of the convolution will be passed through the activation function.[14]
  45. After a convolution layer, it is common to add a pooling layer in between CNN layers.[14]
  46. Training a CNN works in the same way as a regular neural network, using backpropagration or gradient descent.[14]
  47. A convolutional neural network, or CNN, is a deep learning neural network designed for processing structured arrays of data such as images.[15]
  48. The architecture of a convolutional neural network is a multi-layered feed-forward neural network, made by stacking many hidden layers on top of each other in sequence.[15]
  49. A simple convolutional neural network that aids understanding of the core design principles is the early convolutional neural network LeNet-5, published by Yann LeCun in 1998.[15]
  50. Typically, the first layer of a convolutional neural network contains a vertical line detector, a horizontal line detector, and various diagonal, curve and corner detectors.[15]
  51. The neurons within a CNN are split into a three-dimensional structure, with each set of neurons analyzing a small region or feature of the image.[16]
  52. A Convolutional Neural Network (CNN) is a deep learning algorithm that can recognize and classify features in images for computer vision.[16]
  53. The architecture of a CNN is a key factor in determining its performance and efficiency.[16]
  54. The ImageNet project has more than 14 million images specifically designed for training CNN in object detection, one million of which also provide bounding boxes for the use of networks such as YOLO.[16]
  55. Our study proposes a prediction method using one-dimensional convolutional neural network (1-D CNN) that contains all aforementioned processes together.[17]
  56. With the advent of deep learning, 1-D CNN has become favorable for extracting features from time series signals, and therefore for detection, prediction, and classification10.[17]
  57. Therefore, CNN possesses the capacity to extract features from the 1-D time series data of raw ECG signals and use them to monitor mental stress and detect myocardial infractions (MI)11.[17]
  58. We propose a prediction method for VT and VF based on a 1-D CNN trained using HRV signals.[17]
  59. When you input an image into a ConvNet, each of its layers generates several activation maps.[18]
  60. The first (or bottom) layer of the CNN usually detects basic features such as horizontal, vertical, and diagonal edges.[18]
  61. The operation of multiplying pixel values by weights and summing them is called “convolution” (hence the name convolutional neural network).[18]
  62. A CNN is usually composed of several convolution layers, but it also contains other components.[18]
  63. In this section, we describe a new convolutional neural network (CNN) based on the SMILES notation of compounds.[19]
  64. An overview of our CNN is shown in Fig.[19]
  65. Our CNN transforms the SMILES feature matrix into a low-dimensional feature vector termed the SMILES convolution finger print (SCFP).[19]
  66. CNN has multiple layers consisting of two convolutional and pooling layers with a subsequent global pooling layer.[19]
  67. But what is a convolutional neural network and why has it suddenly become so popular?[20]
  68. In addition to exploring how a convolutional neural network (ConvNet) works, we’ll also look at different architectures of a ConvNet and how we can build an object detection model using YOLO.[20]
  69. There are a number of hyperparameters that we can tweak while building a convolutional network.[20]
  70. Let’s look at how a convolution neural network with convolutional and pooling layer works.[20]
  71. A Convolutional Neural Network is a class of artificial neural network that uses convolutional layers to filter inputs for useful information.[21]
  72. The CNN would filter information about the shape of an object when confronted with a general object recognition task but would extract the color of the bird when faced with a bird recognition task.[21]
  73. A convolutional network is different than a regular neural network in that the neurons in its layers are arranged in three dimensions (width, height, and depth dimensions).[21]
  74. This allows the CNN to transform an input volume in three dimensions to an output volume.[21]
  75. Each kernel can output a feature map and all the feature maps are concatenated together, this is also known as a convolutional layer and it is the core component in a CNN.[22]
  76. A typical CNN is usually structured as a series of layers, including multiple convolutional layers and a few of fully connected layers.[22]
  77. Therefore, the process of training a CNN model is transformed into an optimization problem, which normally seeks to minimize the value of the loss function over the training data.[22]
  78. Specifically, a gradient-descent based algorithm is usually adopted to iteratively optimize the parameters in a CNN.[22]
  79. A schematic of a two-layer CNN architecture is shown in Fig.[23]
  80. such that the CNN performs the task that is required.[23]
  81. The main aim of this work is to investigate the applicability of CNN in impact detection and identification of complex composite structures such as aircraft stiffened panel.[24]
  82. Thus, using a CNN with raw data as input will be more advantageous than traditional extraction methods.[24]
  83. It is used to normalise the output vector of the CNN, which is of length equal to the number of classes, say, to a vector of length, whose values sum to 1.[24]
  84. The nodes at the output of the layer, will, thus, contain the probabilities of the input to the CNN belonging to all classes.[24]
  85. In this post I am going to explain what they are and how you can create a convolutional neural network in Keras with Python.[25]
  86. We (along with many CNN implementations) are technically actually using cross-correlation instead of convolution here, but they do almost the same thing.[26]
  87. A CNN trained on MNIST might look for the digit 1, for example, by using an edge-detection filter and checking for two prominent vertical edges near the center of the image.[26]
  88. For our MNIST CNN, we’ll use a small conv layer with 8 filters as the initial layer in our network.[26]
  89. For our MNIST CNN, we’ll place a Max Pooling layer with a pool size of 2 right after our initial conv layer.[26]
  90. We will conclude the chapter with a full working example of LeNet, the first convolutional network successfully deployed, long before the rise of modern deep learning.[27]
  91. To solve the abovementioned two problems, this paper proposes a fault-diagnosis algorithm based on “end-to-end” one-dimensional convolutional neural network.[28]
  92. The model does not rely on artificial feature extraction and expert knowledge and maximizes the use of CNN for feature extraction.[28]
  93. Section 2 gives a brief introduction of CNN.[28]
  94. The CNN is a special structure of the feed-forward neural network, generally including a convolution layer, activation layer, pooling layer, and fully connected layer.[28]

소스

  1. ↑ 1.0 1.1 1.2 1.3 CS231n Convolutional Neural Networks for Visual Recognition
  2. ↑ Convolutional Neural Network
  3. ↑ 3.0 3.1 3.2 Convolutional Neural Network (CNN)
  4. ↑ Understanding of Convolutional Neural Network (CNN) — Deep Learning
  5. ↑ 5.0 5.1 5.2 5.3 A Comprehensive Guide to Convolutional Neural Networks — the ELI5 way
  6. ↑ 6.0 6.1 6.2 6.3 Convolutional Neural Network - an overview
  7. ↑ 7.0 7.1 7.2 7.3 Convolutional neural networks: an overview and application in radiology
  8. ↑ 8.0 8.1 How Do Convolutional Layers Work in Deep Learning Neural Networks?
  9. ↑ 9.0 9.1 9.2 9.3 Convolutional neural network
  10. ↑ 10.0 10.1 10.2 10.3 ML Practicum: Image Classification
  11. ↑ 11.0 11.1 11.2 11.3 An Intuitive Explanation of Convolutional Neural Networks
  12. ↑ 12.0 12.1 12.2 12.3 Convolutional neural network: a review of models, methodologies and applications to object detection
  13. ↑ 13.0 13.1 13.2 A Beginner's Guide to Convolutional Neural Networks (CNNs)
  14. ↑ 14.0 14.1 14.2 14.3 An intuitive guide to Convolutional Neural Networks
  15. ↑ 15.0 15.1 15.2 15.3 Convolutional Neural Network
  16. ↑ 16.0 16.1 16.2 16.3 Convolutional Neural Network Architecture: Forging Pathways to the Future
  17. ↑ 17.0 17.1 17.2 17.3 Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features
  18. ↑ 18.0 18.1 18.2 18.3 What are convolutional neural networks (CNN)?
  19. ↑ 19.0 19.1 19.2 19.3 Convolutional neural network based on SMILES representation of compounds for detecting chemical motif
  20. ↑ 20.0 20.1 20.2 20.3 Tutorial On Convolutional Neural Networks
  21. ↑ 21.0 21.1 21.2 21.3 Convolutional Neural Network (CNN)
  22. ↑ 22.0 22.1 22.2 22.3 Advances in Convolutional Neural Networks
  23. ↑ 23.0 23.1 A mixed-scale dense convolutional neural network for image analysis
  24. ↑ 24.0 24.1 24.2 24.3 A Convolutional Neural Network for Impact Detection and Characterization of Complex Composite Structures
  25. ↑ How to build a convolutional neural network in Keras
  26. ↑ 26.0 26.1 26.2 26.3 CNNs, Part 1: An Introduction to Convolutional Neural Networks
  27. ↑ 6. Convolutional Neural Networks — Dive into Deep Learning 0.15.1 documentation
  28. ↑ 28.0 28.1 28.2 28.3 Multitask Convolutional Neural Network for Rolling Element Bearing Fault Identification

메타데이터

위키데이터

Spacy 패턴 목록

  • [{'LOWER': 'convolutional'}, {'LOWER': 'neural'}, {'LEMMA': 'network'}]
  • [{'LEMMA': 'CNN'}]
  • [{'LEMMA': 'ConvNet'}]
  • [{'LOWER': 'convolutional'}, {'LEMMA': 'Network'}]
  • [{'LOWER': 'convolution'}, {'LOWER': 'neural'}, {'LEMMA': 'Network'}]
  • [{'LOWER': 'convolutional'}, {'LEMMA': 'net'}]
원본 주소 "https://wiki.mathnt.net/index.php?title=합성곱_신경망&oldid=51638"