[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-572641-105":3,"detail-sidebar-cat-1-en-105":81,"doc-detail-572641-en":127},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","super-vip-cheatsheet-deep-learning-overview-and-key-concepts","Super VIP Cheatsheet - Deep Learning - Overview and Key Concepts","","Super VIP Cheatsheet: Deep Learning provides a structured, sectioned reference for core deep learning topics. It covers convolutional neural networks, including common layer types, filter hyperparameters, stride and zero-padding, and how tuning choices affect output feature-map size. It also outlines recurrent neural networks for long-term dependencies, word representation, language modeling, and attention. Additional tips focus on data processing, training, parameter tuning, regularization, and good practices.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/template/","Template",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/template/presentations/","Presentations",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/template/super-vip-cheatsheet-deep-learning-overview-and-key-concepts/572641/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/super-vip-cheatsheet-deep-learning-overview-and-key-concepts/572641.png","ImageObject",442,249,{"name":42,"@type":43},"iuabcfuw","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-21","2026-10-13",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",6,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What does the cheatsheet cover for convolutional neural networks (CNNs)?","Question",{"text":63,"@type":64},"It explains CNN overview, the main layer types (convolution, pooling, and fully connected), and key filter hyperparameters such as dimensions, stride, and zero-padding.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How do stride and zero-padding affect CNN output?",{"text":68,"@type":64},"Stride determines how the window moves across the input, while zero-padding adds values around the borders to control feature-map sizing and preserve or drop boundary computations.",{"name":70,"@type":61,"acceptedAnswer":71},"Which recurrent neural network (RNN) topics are included?",{"text":72,"@type":64},"It covers handling long-term dependencies, learning word representations (including word embeddings), comparing words, language modeling, machine translation, and attention.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},572641,1792051017,{"code":4,"msg":82,"data":83},"success",[84,88,93,98,103,108,113,118,123],{"id":85,"doc_module":22,"doc_module_name":25,"category_name":29,"show_sort_weight":86,"slug":87},11,90,"presentations",{"id":89,"doc_module":22,"doc_module_name":25,"category_name":90,"show_sort_weight":91,"slug":92},12,"Resumes",80,"resumes",{"id":94,"doc_module":22,"doc_module_name":25,"category_name":95,"show_sort_weight":96,"slug":97},14,"Invoices",70,"invoices",{"id":99,"doc_module":22,"doc_module_name":25,"category_name":100,"show_sort_weight":101,"slug":102},15,"Posters",60,"posters",{"id":104,"doc_module":22,"doc_module_name":25,"category_name":105,"show_sort_weight":106,"slug":107},16,"Social Media",50,"social-media",{"id":109,"doc_module":22,"doc_module_name":25,"category_name":110,"show_sort_weight":111,"slug":112},17,"Forms",40,"forms",{"id":114,"doc_module":22,"doc_module_name":25,"category_name":115,"show_sort_weight":116,"slug":117},18,"Letters",30,"letters",{"id":119,"doc_module":22,"doc_module_name":25,"category_name":120,"show_sort_weight":121,"slug":122},21,"Paper Templates",5,"papers-templates",{"id":124,"doc_module":22,"doc_module_name":25,"category_name":125,"show_sort_weight":4,"slug":126},158,"General","general-158",{"code":4,"msg":82,"data":128},{"doc_id":79,"user_id":129,"nickname":42,"user_avatar":130,"doc_module":22,"category_id":85,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":131,"file_id":132,"file_url":133,"file_type":134,"file_size":135,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":136,"language":137,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":138,"faqs":139,"seo_title":140,"seo_description":12,"update_tm":141,"read_time":104},3985747874027,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","Super VIP Cheatsheet: Deep Learning  \nAfshine Amidi and Shervine Amidi  \nNovember 25, 2018  \nContents  \n1 Convolutional Neural Networks 2  \n1.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2  \n1.2 Types of layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2  \n1.3 Filter hyperparameters . . . . . . . . . . . . . . . . . . . . . . . . . . 2  \n1.4 Tuning hyperparameters . . . . . . . . . . . . . . . . . . . . . . . . . 3  \n1.5 Commonly used activation functions . . . . . . . . . . . . . . . . . . . 3  \n1.6 Object detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4  \n1.6.1 Face verification and recognition . . . . . . . . . . . . . . . . . 5  \n1.6.2 Neural style transfer . . . . . . . . . . . . . . . . . . . . . . . 5  \n1.6.3 Architectures using computational tricks . . . . . . . . . . . . 6  \n2 Recurrent Neural Networks 7  \n2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7  \n2.2 Handling long term dependencies . . . . . . . . . . . . . . . . . . . . 8  \n2.3 Learning word representation . . . . . . . . . . . . . . . . . . . . . . 9  \n2.3. 1 Motivation and notations . . . . . . . . . . . . . . . . . . . 9  \n2.3.2 Word embeddings . . . . . . . . . . . . . . . . . . . . . . . 9  \n2.4 Comparing words . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9  \n2.5 Language model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10  \n2.6 Machine translation . . . . . . . . . . . . . . . . . . . . . . . . . . . 10  \n2.7 Attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10  \n3 Deep Learning Tips and Tricks 11  \n3.1 Data processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11  \n3.2 Training a neural network . . . . . . . . . . . . . . . . . . . . . . . . 12  \n3.2.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12  \n3.2.2 Finding optimal weights . . . . . . . . . . . . . . . . . . . . . 12  \n3.3 Parameter tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12  \n3.3.1 Weights initialization . . . . . . . . . . . . . . . . . . . . . . 12  \n3.3.2 Optimizing convergence . . . . . . . . . . . . . . . . . . . . . 12  \n3.4 Regularization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13  \n3.5 Good practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13  \n1 Convolutional Neural Networks  \n1.1 Overview  \nr Architecture of a traditional CNN – Convolutional neural networks, also known as CNNs, are a specific type of neural networks that are generally composed of the following layers:  \nThe convolution layer and the pooling layer can be fine-tuned with respect to hyperparameters that are described in the next sections.  \n1.2 Types of layer  \nr Convolutional layer (CONV)– The convolution layer (CONV) uses filters that perform convolution operations as it is scanning the input I with respect to its dimensions. Its hyperparameters include the filter size F and stride S. The resulting output O is called feature map or  \nactivation map.  \nRemark: the convolution step can be generalized to the 1D and 3D cases as well.  \nr Pooling (POOL)– The pooling layer (POOL) is a downsampling operation, typically applied after a convolution layer, which does some spatial invariance. In particular, max and average pooling are special kinds of pooling where the maximum and average value is taken, respectively.  \n\n|  | Max pooling | Average pooling |\n| --- | --- | --- |\n| Purpose | Each pooling operation selects the maximum value of the current view | Each pooling operation averages the values of the current view |\n| Illustration | | |\n| Comments | -Preserves detected features\u003Cbr>-Most commonly used | -Downsamples feature map\u003Cbr>-Used in LeNet |\n\nr Fully Connected (FC)– The fully connected layer (FC) operates on a flattened input where each input is connected to all neurons. If present, FC layers are usually found towards the end of CNN architectures and can be used to optimiz","cbCaikMFIUd1ONFp","https://ap.wps.com/l/cbCaikMFIUd1ONFp","pdf",7465206,47,"English","# Convolutional Neural Networks\n## Overview\n## Types of layer\n## Filter hyperparameters\n## Tuning hyperparameters\n## Commonly used activation functions\n## Object detection\n# Recurrent Neural Networks\n## Overview\n## Handling long term dependencies\n## Learning word representation\n## Comparing words\n## Language model\n## Machine translation\n## Attention\n# Deep Learning Tips and Tricks\n## Data processing\n## Training a neural network\n## Parameter tuning\n## Regularization\n## Good practices","[{\"question\":\"What does the cheatsheet cover for convolutional neural networks (CNNs)?\",\"answer\":\"It explains CNN overview, the main layer types (convolution, pooling, and fully connected), and key filter hyperparameters such as dimensions, stride, and zero-padding.\"},{\"question\":\"How do stride and zero-padding affect CNN output?\",\"answer\":\"Stride determines how the window moves across the input, while zero-padding adds values around the borders to control feature-map sizing and preserve or drop boundary computations.\"},{\"question\":\"Which recurrent neural network (RNN) topics are included?\",\"answer\":\"It covers handling long-term dependencies, learning word representations (including word embeddings), comparing words, language modeling, machine translation, and attention.\"}]","Super VIP Cheatsheet - Deep Learning - Overview and Key Concepts | PDF",1791928588]