[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-instillai--deep-learning-roadmap":3,"tool-instillai--deep-learning-roadmap":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":79,"owner_website":81,"owner_url":82,"languages":83,"stars":88,"forks":89,"last_commit_at":90,"license":91,"difficulty_score":92,"env_os":93,"env_gpu":94,"env_ram":94,"env_deps":95,"category_tags":98,"github_topics":99,"view_count":23,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":102,"updated_at":103,"faqs":104,"releases":105},3848,"instillai\u002Fdeep-learning-roadmap","deep-learning-roadmap",":satellite: All You Need to Know About Deep Learning - A kick-starter","deep-learning-roadmap 是一个专为深度学习爱好者打造的开源资源导航项目，旨在为开发者和研究人员提供一条清晰的学习捷径。面对网络上浩如烟海且分散的深度学习资料，初学者往往难以辨别重点，资深从业者也常耗费大量时间在资源检索上。该项目通过高度结构化的分类体系，将海量的论文、教程和工具整理得井井有条，让用户能根据具体需求快速定位到最相关的核心内容。\n\n即使你对学习路径尚感迷茫，项目中提供的通用基础资源也能帮助你顺利起步。其独特的亮点在于“精准靶向”的资源组织方式，不仅覆盖了从入门到进阶的各类主题，还特别设立了专门的论文章节，方便用户追踪前沿学术动态。此外，项目还关联了免费的机器学习电子书下载及交流社区，进一步丰富了学习生态。无论是想要系统构建知识体系的在校学生，还是希望高效查找特定技术资料的算法工程师，都能在这里找到宝贵的指引，让深度学习之旅变得更加高效和轻松。","###################################################\nDeep Learning - All You Need to Know\n###################################################\n\n.. image:: https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fcontributions-welcome-brightgreen.svg?style=flat\n    :target: https:\u002F\u002Fgithub.com\u002Fosforscience\u002Fdeep-learning-all-you-need\u002Fpulls\n.. image:: https:\u002F\u002Fbadges.frapsoft.com\u002Fos\u002Fv2\u002Fopen-source.png?v=103\n    :target: https:\u002F\u002Fgithub.com\u002Fellerbrock\u002Fopen-source-badge\u002F\n.. image:: https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fl\u002Fansicolortags.svg\n      :target: https:\u002F\u002Fgithub.com\u002Fosforscience\u002Fdeep-learning-all-you-need\u002Fblob\u002Fmaster\u002FLICENSE\n.. image:: https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Fmachinemindset.svg?label=Follow&style=social\n      :target: https:\u002F\u002Ftwitter.com\u002Fmachinemindset\n      \n##########################################################################\nSponsorship\n##########################################################################\n\nTo support maintaining and upgrading this project, please kindly consider `Sponsoring the project developer \u003Chttps:\u002F\u002Fgithub.com\u002Fsponsors\u002Fastorfi\u002Fdashboard>`_.\n\nAny level of support is a great contribution here :heart:\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsponsors\u002Fastorfi\u002Fdashboard\" target=\"_blank\">\n  \u003Cimg width=\"600\" height=\"500\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_f47972565d7a.jpg\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>      \n      \n\n###################################################\nDownload Free Python Machine Learning Book\n###################################################\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"http:\u002F\u002Fwww.machinelearningmindset.com\u002Fdeep-learning-roadmap\u002F\" target=\"_blank\">\n  \u003Cimg width=\"900\" height=\"625\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_acaf49a52074.png\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>\n   \n   \n###################################################\nSlack Group\n###################################################\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"https:\u002F\u002Fwww.machinelearningmindset.com\u002Fslack-group\u002F\" target=\"_blank\">\n  \u003Cimg width=\"1033\" height=\"350\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_2f07fac72f26.png\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>\n\n\n##################\nTable of Contents\n##################\n.. contents::\n  :local:\n  :depth: 4\n\n.. image:: _img\u002Fmainpage\u002Flogo.gif\n\n***************\nIntroduction\n***************\n\nThe purpose of this project is to introduce a shortcut to developers and researcher\nfor finding useful resources about Deep Learning.\n\n============\nMotivation\n============\n\nThere are different motivations for this open source project.\n\n.. --------------------\n.. Why Deep Learning?\n.. --------------------\n\n------------------------------------------------------------\nWhat's the point of this open source project?\n------------------------------------------------------------\n\nThere are other repositories similar to this repository that are very\ncomprehensive and useful and to be honest they made me ponder if there is\na necessity for this repository!\n\n**The point of this repository is that the resources are being targeted**. The organization\nof the resources is such that the user can easily find the things he\u002Fshe is looking for.\nWe divided the resources to a large number of categories that in the beginning one may\nhave a headache!!! However, if someone knows what is being located, it is very easy to find the most related resources.\nEven if someone doesn't know what to look for, in the beginning, the general resources have\nbeen provided.\n\n\n.. ================================================\n.. How to make the most of this effort\n.. ================================================\n\n************\nPapers\n************\n\n.. image:: _img\u002Fmainpage\u002Farticle.jpeg\n\nThis chapter is associated with the papers published in deep learning.\n\n====================\nModels\n====================\n\n-----------------------\nConvolutional Networks\n-----------------------\n\n  .. image:: _img\u002Fmainpage\u002Fconvolutional.png\n\n.. For continuous lines, the lines must be start from the same locations.\n* **Imagenet classification with deep convolutional neural networks** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4824-imagenet-classification-with-deep-convolutional-neural-networks>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fdontfollowmeimcrazy\u002Fimagenet>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Convolutional Neural Networks for Sentence Classification** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1408.5882>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyoonkim\u002FCNN_sentence>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Large-scale Video Classification with Convolutional Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FKarpathy_Large-scale_Video_Classification_2014_CVPR_paper.html>`_][`Project Page \u003Chttps:\u002F\u002Fcs.stanford.edu\u002Fpeople\u002Fkarpathy\u002Fdeepvideo\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Learning and Transferring Mid-Level Image Representations using Convolutional Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FOquab_Learning_and_Transferring_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n* **Deep convolutional neural networks for LVCSR** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6639347\u002F&hl=zh-CN&sa=T&oi=gsb&ct=res&cd=0&ei=KknXWYbGFMbFjwSsyICADQ&scisig=AAGBfm2F0Zlu0ciUwadzshNNm80IQQhuhA>`_]\n  \n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Face recognition: a convolutional neural-network approach** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F554195\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n\n-----------------------\nRecurrent Networks\n-----------------------\n\n  .. image:: _img\u002Fmainpage\u002FRecurrent_neural_network_unfold.svg\n\n\n.. For continuous lines, the lines must be start from the same locations.\n* **An empirical exploration of recurrent network architectures** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fjozefowicz15.pdf?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=revue>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fdebajyotidatta\u002FRecurrentArchitectures>`_]\n\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **LSTM: A search space odyssey** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F7508408\u002F>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Ffomorians\u002Flstm-odyssey>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n* **On the difficulty of training recurrent neural networks** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv28\u002Fpascanu13.pdf>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fpascanur\u002FtrainingRNNs>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Learning to forget: Continual prediction with LSTM** :\n  [`Paper \u003Chttp:\u002F\u002Fdigital-library.theiet.org\u002Fcontent\u002Fconferences\u002F10.1049\u002Fcp_19991218>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n-----------------------\nAutoencoders\n-----------------------\n\n.. image:: _img\u002Fmainpage\u002FAutoencoder_structure.png\n\n* **Extracting and composing robust features with denoising autoencoders** :\n  [`Paper \u003Chttps:\u002F\u002Fdl.acm.org\u002Fcitation.cfm?id=1390294>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion** :\n  [`Paper \u003Chttp:\u002F\u002Fwww.jmlr.org\u002Fpapers\u002Fv11\u002Fvincent10a.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Frajarsheem\u002Flibsdae-autoencoder-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Adversarial Autoencoders** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.05644>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fconan7882\u002Fadversarial-autoencoders>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Autoencoders, Unsupervised Learning, and Deep Architectures** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv27\u002Fbaldi12a\u002Fbaldi12a.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Reducing the Dimensionality of Data with Neural Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fscience.sciencemag.org\u002Fcontent\u002F313\u002F5786\u002F504>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fjordn\u002Fautoencoder>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n-----------------------\nGenerative Models\n-----------------------\n\n.. image:: _img\u002Fmainpage\u002Fgenerative.png\n\n* **Exploiting generative models discriminative classifiers** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F1520-exploiting-generative-models-in-discriminative-classifiers.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Semi-supervised Learning with Deep Generative Models** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5352-semi-supervised-learning-with-deep-generative-models>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fwohlert\u002Fsemi-supervised-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n* **Generative Adversarial Nets** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5423-generative-adversarial-nets>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fgoodfeli\u002Fadversarial>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Generalized Denoising Auto-Encoders as Generative Models** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5023-generalized-denoising-auto-encoders-as-generative-models>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n  \n* **Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06434>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002FDCGAN-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n-----------------------\nProbabilistic Models\n-----------------------\n\n* **Stochastic Backpropagation and Approximate Inference in Deep Generative Models** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1401.4082>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Probabilistic models of cognition: exploring representations and inductive biases** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1364661310001129>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **On deep generative models with applications to recognition** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F5995710\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n\n\n\n====================\nCore\n====================\n\n---------------------\nOptimization\n---------------------\n\n.. ################################################################################\n.. For continuous lines, the lines must be start from the same locations.\n* **Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1502.03167>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Dropout: A Simple Way to Prevent Neural Networks from Overfitting** :\n  [`Paper \u003Chttp:\u002F\u002Fwww.jmlr.org\u002Fpapers\u002Fvolume15\u002Fsrivastava14a\u002Fsrivastava14a.pdf?utm_content=buffer79b43&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Training Very Deep Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5850-training-very-deep-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FHe_Delving_Deep_into_ICCV_2015_paper.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Large Scale Distributed Deep Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4687-large-scale-distributed-deep-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n------------------------\nRepresentation Learning\n------------------------\n\n* **Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06434>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002FNewmu\u002Fdcgan_code>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Representation Learning: A Review and New Perspectives** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6472238\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F6399-infogan-interpretable-representation>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fopenai\u002FInfoGAN>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n------------------------------------\nUnderstanding and Transfer Learning\n------------------------------------\n\n* **Learning and Transferring Mid-Level Image Representations using Convolutional Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FOquab_Learning_and_Transferring_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Distilling the Knowledge in a Neural Network** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1503.02531>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv32\u002Fdonahue14.pdf>`_][\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **How transferable are features in deep neural networks?** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5347-how-transferable-are-features-in-deep-n%E2%80%A6>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyosinski\u002Fconvnet_transfer>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n-----------------------\nReinforcement Learning\n-----------------------\n\n* **Human-level control through deep reinforcement learning** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.nature.com\u002Farticles\u002Fnature14236\u002F>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fdevsisters\u002FDQN-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Playing Atari with Deep Reinforcement Learning** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.5602>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002Fdeep-rl-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Continuous control with deep reinforcement learning** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1509.02971>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fstevenpjg\u002Fddpg-aigym>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Deep Reinforcement Learning with Double Q-Learning** :\n  [`Paper \u003Chttp:\u002F\u002Fwww.aaai.org\u002Focs\u002Findex.php\u002FAAAI\u002FAAAI16\u002Fpaper\u002Fdownload\u002F12389\u002F11847>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002Fdeep-rl-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Dueling Network Architectures for Deep Reinforcement Learning** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06581>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyoosan\u002Fdeeprl>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n====================\nApplications\n====================\n\n--------------------\nImage Recognition\n--------------------\n\n* **Deep Residual Learning for Image Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fhtml\u002FHe_Deep_Residual_Learning_CVPR_2016_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fgcr\u002Ftorch-residual-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Very Deep Convolutional Networks for Large-Scale Image Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1409.1556>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Multi-column Deep Neural Networks for Image Classification** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1202.2745>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **DeepID3: Face Recognition with Very Deep Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1502.00873>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.6034>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fartvandelay\u002FDeep_Inside_Convolutional_Networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Deep Image: Scaling up Image Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fvc\u002Farxiv\u002Fpapers\u002F1501\u002F1501.02876v1.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Long-Term Recurrent Convolutional Networks for Visual Recognition and Description** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FDonahue_Long-Term_Recurrent_Convolutional_2015_CVPR_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002FJaggerYoung\u002FLRCN-for-Activity-Recognition>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **3D Convolutional Neural Networks for Cross Audio-Visual Matching Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F8063416>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fastorfi\u002Flip-reading-deeplearning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n--------------------\nObject Recognition\n--------------------\n\n* **ImageNet Classification with Deep Convolutional Neural Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4824-imagenet-classification-with-deep-convolutional-neural-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Learning Deep Features for Scene Recognition using Places Database** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5349-learning-deep-features>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Scalable Object Detection using Deep Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FErhan_Scalable_Object_Detection_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5638-faster-r-cnn-towards-real-time-object-detection-with-region-proposal-networks>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Frbgirshick\u002Fpy-faster-rcnn>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.6229>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fsermanet\u002FOverFeat>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **CNN Features Off-the-Shelf: An Astounding Baseline for Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_workshops_2014\u002FW15\u002Fhtml\u002FRazavian_CNN_Features_Off-the-Shelf_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **What is the best multi-stage architecture for object recognition?** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F5459469\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_2.png\n\n\n--------------------\nAction Recognition\n--------------------\n\n* **Long-Term Recurrent Convolutional Networks for Visual Recognition and Description** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FDonahue_Long-Term_Recurrent_Convolutional_2015_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Learning Spatiotemporal Features With 3D Convolutional Networks** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fhtml\u002FTran_Learning_Spatiotemporal_Features_ICCV_2015_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002FDavideA\u002Fc3d-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Describing Videos by Exploiting Temporal Structure** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fhtml\u002FYao_Describing_Videos_by_ICCV_2015_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Ftsenghungchen\u002FSA-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Convolutional Two-Stream Network Fusion for Video Action Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fhtml\u002FFeichtenhofer_Convolutional_Two-Stream_Network_CVPR_2016_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Ffeichtenhofer\u002Ftwostreamfusion>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Temporal segment networks: Towards good practices for deep action recognition** :\n  [`Paper \u003Chttps:\u002F\u002Flink.springer.com\u002Fchapter\u002F10.1007\u002F978-3-319-46484-8_2>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyjxiong\u002Ftemporal-segment-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n----------------------------\nCaption Generation\n----------------------------\n\n* **Show, Attend and Tell: Neural Image Caption Generation with Visual Attention** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fxuc15.pdf>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyunjey\u002Fshow-attend-and-tell>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Mind's Eye: A Recurrent Visual Representation for Image Caption Generation** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FChen_Minds_Eye_A_2015_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_2.png\n\n* **Generative Adversarial Text to Image Synthesis** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Freed16.pdf>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fzsdonghao\u002Ftext-to-image>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Deep Visual-Semantic Al60ignments for Generating Image Descriptions** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FKarpathy_Deep_Visual-Semantic_Alignments_2015_CVPR_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fjonkuo\u002FDeep-Learning-Image-Captioning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Show and Tell: A Neural Image Caption Generator** :\n  [`Paper \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FVinyals_Show_and_Tell_2015_CVPR_paper.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002FDeepRNN\u002Fimage_captioning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n----------------------------\nNatural Language Processing\n----------------------------\n\n* **Distributed Representations of Words and Phrases and their Compositionality** :\n  [`Paper \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf>`_][`Code \u003Chttps:\u002F\u002Fcode.google.com\u002Farchive\u002Fp\u002Fword2vec\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Efficient Estimation of Word Representations in Vector Space** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1301.3781.pdf>`_][`Code \u003Chttps:\u002F\u002Fcode.google.com\u002Farchive\u002Fp\u002Fword2vec\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Sequence to Sequence Learning with Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1409.3215.pdf>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Ffarizrahman4u\u002Fseq2seq>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Neural Machine Translation by Jointly Learning to Align and Translate** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1409.0473.pdf>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fnmt>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Get To The Point: Summarization with Pointer-Generator Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1704.04368>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fabisee\u002Fpointer-generator>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Attention Is All You Need** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fjadore801120\u002Fattention-is-all-you-need-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Convolutional Neural Networks for Sentence Classification** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1408.5882>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fyoonkim\u002FCNN_sentence>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n----------------------------\nSpeech Technology\n----------------------------\n\n* **Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6296526\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Towards End-to-End Speech Recognition with Recurrent Neural Networks** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv32\u002Fgraves14.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Speech recognition with deep recurrent neural networks** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6638947\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Fast and Accurate Recurrent Neural Network Acoustic Models for Speech Recognition** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1507.06947>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Deep Speech 2 : End-to-End Speech Recognition in English and Mandarin** :\n  [`Paper \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Famodei16.html>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002FDeepSpeech>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **A novel scheme for speaker recognition using a phonetically-aware deep neural network** :\n  [`Paper \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6853887\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n \n* **Text-Independent Speaker Verification Using 3D Convolutional Neural Networks** :\n  [`Paper \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1705.09422>`_][`Code \u003Chttps:\u002F\u002Fgithub.com\u002Fastorfi\u002F3D-convolutional-speaker-recognition>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n************\nDatasets\n************\n\n====================\nImage\n====================\n\n\n----------------------------\nGeneral\n----------------------------\n\n* **MNIST** Handwritten digits:\n  [`Link \u003Chttp:\u002F\u002Fyann.lecun.com\u002Fexdb\u002Fmnist\u002F>`_]\n\n\n----------------------------\nFace\n----------------------------\n\n* **Face Recognition Technology (FERET)** The goal of the FERET program was to develop automatic face recognition capabilities that could be employed to assist security, intelligence, and law enforcement personnel in the performance of their duties:\n  [`Link \u003Chttps:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-technology-feret>`_]\n\n* **The CMU Pose, Illumination, and Expression (PIE) Database of Human Faces** Between October and December 2000 we collected a database of 41,368 images of 68 people:\n  [`Link \u003Chttps:\u002F\u002Fwww.ri.cmu.edu\u002Fpublications\u002Fthe-cmu-pose-illumination-and-expression-pie-database-of-human-faces\u002F>`_]\n\n* **YouTube Faces DB** The data set contains 3,425 videos of 1,595 different people. All the videos were downloaded from YouTube. An average of 2.15 videos are available for each subject:\n  [`Link \u003Chttps:\u002F\u002Fwww.cs.tau.ac.il\u002F~wolf\u002Fytfaces\u002F>`_]\n\n* **Grammatical Facial Expressions Data Set** Developed to assist the the automated analysis of facial expressions:\n  [`Link \u003Chttps:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FGrammatical+Facial+Expressions>`_]\n\n* **FaceScrub** A Dataset With Over 100,000 Face Images of 530 People:\n  [`Link \u003Chttp:\u002F\u002Fvintage.winklerbros.net\u002Ffacescrub.html>`_]\n\n* **IMDB-WIKI** 500k+ face images with age and gender labels:\n  [`Link \u003Chttps:\u002F\u002Fdata.vision.ee.ethz.ch\u002Fcvl\u002Frrothe\u002Fimdb-wiki\u002F>`_]\n\n* **FDDB** Face Detection Data Set and Benchmark (FDDB):\n  [`Link \u003Chttp:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002F>`_]\n\n----------------------------\nObject Recognition\n----------------------------\n\n* **COCO** Microsoft COCO: Common Objects in Context:\n  [`Link \u003Chttp:\u002F\u002Fcocodataset.org\u002F#home>`_]\n\n* **ImageNet** The famous ImageNet dataset:\n  [`Link \u003Chttp:\u002F\u002Fwww.image-net.org\u002F>`_]\n\n* **Open Images Dataset** Open Images is a dataset of ~9 million images that have been annotated with image-level labels and object bounding boxes:\n  [`Link \u003Chttps:\u002F\u002Fstorage.googleapis.com\u002Fopenimages\u002Fweb\u002Findex.html>`_]\n\n* **Caltech-256 Object Category Dataset** A large dataset object classification:\n  [`Link \u003Chttps:\u002F\u002Fauthors.library.caltech.edu\u002F7694\u002F>`_]\n\n* **Pascal VOC dataset** A large dataset for classification tasks:\n  [`Link \u003Chttp:\u002F\u002Fhost.robots.ox.ac.uk\u002Fpascal\u002FVOC\u002F>`_]\n\n* **CIFAR 10 \u002F CIFAR 100** The CIFAR-10 dataset consists of 60000 32x32 colour images in 10 classes. CIFAR-100 is similar to CIFAR-10 but it has 100 classes containing 600 images each:\n  [`Link \u003Chttps:\u002F\u002Fwww.cs.toronto.edu\u002F~kriz\u002Fcifar.html>`_]\n\n\n----------------------------\nAction recognition\n----------------------------\n\n* **HMDB** a large human motion database:\n  [`Link \u003Chttp:\u002F\u002Fserre-lab.clps.brown.edu\u002Fresource\u002Fhmdb-a-large-human-motion-database\u002F>`_]\n\n* **MHAD** Berkeley Multimodal Human Action Database:\n  [`Link \u003Chttp:\u002F\u002Ftele-immersion.citris-uc.org\u002Fberkeley_mhad>`_]\n\n* **UCF101 - Action Recognition Data Set** UCF101 is an action recognition data set of realistic action videos, collected from YouTube, having 101 action categories. This data set is an extension of UCF50 data set which has 50 action categories:\n  [`Link \u003Chttp:\u002F\u002Fcrcv.ucf.edu\u002Fdata\u002FUCF101.php>`_]\n\n* **THUMOS Dataset** A large dataset for action classification:\n  [`Link \u003Chttp:\u002F\u002Fcrcv.ucf.edu\u002Fdata\u002FTHUMOS.php>`_]\n\n* **ActivityNet** A Large-Scale Video Benchmark for Human Activity Understanding:\n  [`Link \u003Chttp:\u002F\u002Factivity-net.org\u002F>`_]\n\n======================================\nText and Natural Language Processing\n======================================\n\n\n-----------------------\nGeneral\n-----------------------\n\n* **1 Billion Word Language Model Benchmark**: The purpose of the project is to make available a standard training and test setup for language modeling experiments:\n  [`Link \u003Chttp:\u002F\u002Fwww.statmt.org\u002Flm-benchmark\u002F>`_]\n\n* **Common Crawl**: The Common Crawl corpus contains petabytes of data collected over the last 7 years. It contains raw web page data, extracted metadata and text extractions:\n  [`Link \u003Chttp:\u002F\u002Fcommoncrawl.org\u002Fthe-data\u002Fget-started\u002F>`_]\n\n* **Yelp Open Dataset**: A subset of Yelp's businesses, reviews, and user data for use in personal, educational, and academic purposes:\n  [`Link \u003Chttps:\u002F\u002Fwww.yelp.com\u002Fdataset>`_]\n\n\n-----------------------\nText classification\n-----------------------\n\n* **20 newsgroups** The 20 Newsgroups data set is a collection of approximately 20,000 newsgroup documents, partitioned (nearly) evenly across 20 different newsgroups:\n  [`Link \u003Chttp:\u002F\u002Fqwone.com\u002F~jason\u002F20Newsgroups\u002F>`_]\n\n* **Broadcast News** The 1996 Broadcast News Speech Corpus contains a total of 104 hours of broadcasts from ABC, CNN and CSPAN television networks and NPR and PRI radio networks with corresponding transcripts:\n  [`Link \u003Chttps:\u002F\u002Fcatalog.ldc.upenn.edu\u002FLDC97S44>`_]\n\n* **The wikitext long term dependency language modeling dataset**: A collection of over 100 million tokens extracted from the set of verified Good and Featured articles on Wikipedia. :\n  [`Link \u003Chttps:\u002F\u002Feinstein.ai\u002Fresearch\u002Fthe-wikitext-long-term-dependency-language-modeling-dataset>`_]\n\n-----------------------\nQuestion Answering\n-----------------------\n\n* **Question Answering Corpus** by Deep Mind and Oxford which is two new corpora of roughly a million news stories with associated queries from the CNN and Daily Mail websites.\n  [`Link \u003Chttps:\u002F\u002Fgithub.com\u002Fdeepmind\u002Frc-data>`_]\n\n* **Stanford Question Answering Dataset (SQuAD)** consisting of questions posed by crowdworkers on a set of Wikipedia articles:\n  [`Link \u003Chttps:\u002F\u002Frajpurkar.github.io\u002FSQuAD-explorer\u002F>`_]\n\n* **Amazon question\u002Fanswer data** contains Question and Answer data from Amazon, totaling around 1.4 million answered questions:\n  [`Link \u003Chttp:\u002F\u002Fjmcauley.ucsd.edu\u002Fdata\u002Famazon\u002Fqa\u002F>`_]\n\n\n\n-----------------------\nSentiment Analysis\n-----------------------\n\n* **Multi-Domain Sentiment Dataset** TThe Multi-Domain Sentiment Dataset contains product reviews taken from Amazon.com from many product types (domains):\n  [`Link \u003Chttp:\u002F\u002Fwww.cs.jhu.edu\u002F~mdredze\u002Fdatasets\u002Fsentiment\u002F>`_]\n\n* **Stanford Sentiment Treebank Dataset** The Stanford Sentiment Treebank is the first corpus with fully labeled parse trees that allows for a complete analysis of the compositional effects of sentiment in language:\n  [`Link \u003Chttps:\u002F\u002Fnlp.stanford.edu\u002Fsentiment\u002F>`_]\n\n* **Large Movie Review Dataset**: This is a dataset for binary sentiment classification:\n  [`Link \u003Chttp:\u002F\u002Fai.stanford.edu\u002F~amaas\u002Fdata\u002Fsentiment\u002F>`_]\n\n\n-----------------------\nMachine Translation\n-----------------------\n\n* **Aligned Hansards of the 36th Parliament of Canada** dataset contains 1.3 million pairs of aligned text chunks:\n  [`Link \u003Chttps:\u002F\u002Fwww.isi.edu\u002Fnatural-language\u002Fdownload\u002Fhansard\u002F>`_]\n\n* **Europarl: A Parallel Corpus for Statistical Machine Translation** dataset extracted from the proceedings of the European Parliament:\n  [`Link \u003Chttp:\u002F\u002Fwww.statmt.org\u002Feuroparl\u002F>`_]\n\n\n-----------------------\nSummarization\n-----------------------\n\n* **Legal Case Reports Data Set** as a textual corpus of 4000 legal cases for automatic summarization and citation analysis.:\n  [`Link \u003Chttps:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FLegal+Case+Reports>`_]\n\n\n======================================\nSpeech Technology\n======================================\n\n* **TIMIT Acoustic-Phonetic Continuous Speech Corpus** The TIMIT corpus of read speech is designed to provide speech data for acoustic-phonetic studies and for the development and evaluation of automatic speech recognition systems:\n  [`Link \u003Chttps:\u002F\u002Fcatalog.ldc.upenn.edu\u002Fldc93s1>`_]\n\n* **LibriSpeech** LibriSpeech is a corpus of approximately 1000 hours of 16kHz read English speech, prepared by Vassil Panayotov with the assistance of Daniel Povey:\n  [`Link \u003Chttp:\u002F\u002Fwww.openslr.org\u002F12\u002F>`_]\n\n* **VoxCeleb** A large scale audio-visual dataset:\n  [`Link \u003Chttp:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvoxceleb\u002F>`_]\n\n* **NIST Speaker Recognition**:\n  [`Link \u003Chttps:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fiad\u002Fmig\u002Fspeaker-recognition>`_]\n\n\n\n\n\n\n************\nCourses\n************\n\n.. image:: _img\u002Fmainpage\u002Fonline.png\n\n* **Machine Learning** by Stanford on Coursera :\n  [`Link \u003Chttps:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fmachine-learning>`_]\n\n* **Neural Networks and Deep Learning** Specialization by Coursera:\n  [`Link \u003Chttps:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fneural-networks-deep-learning>`_]\n\n* **Intro to Deep Learning** by Google:\n  [`Link \u003Chttps:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fdeep-learning--ud730>`_]\n\n* **Introduction to Deep Learning** by CMU:\n  [`Link \u003Chttp:\u002F\u002Fdeeplearning.cs.cmu.edu\u002F>`_]\n\n* **NVIDIA Deep Learning Institute** by NVIDIA:\n  [`Link \u003Chttps:\u002F\u002Fwww.nvidia.com\u002Fen-us\u002Fdeep-learning-ai\u002Feducation\u002F>`_]\n\n* **Convolutional Neural Networks for Visual Recognition** by Stanford:\n  [`Link \u003Chttp:\u002F\u002Fcs231n.stanford.edu\u002F>`_]\n\n* **Deep Learning for Natural Language Processing** by Stanford:\n  [`Link \u003Chttp:\u002F\u002Fcs224d.stanford.edu\u002F>`_]\n\n* **Deep Learning** by fast.ai:\n  [`Link \u003Chttp:\u002F\u002Fwww.fast.ai\u002F>`_]\n\n* **Course on Deep Learning for Visual Computing** by IITKGP:\n  [`Link \u003Chttps:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLuv3GM6-gsE1Biyakccxb3FAn4wBLyfWf>`_]\n\n\n\n\n************\nBooks\n************\n\n.. image:: _img\u002Fmainpage\u002Fbooks.jpg\n\n* **Deep Learning** by Ian Goodfellow:\n  [`Link \u003Chttp:\u002F\u002Fwww.deeplearningbook.org\u002F>`_]\n\n* **Neural Networks and Deep Learning** :\n  [`Link \u003Chttp:\u002F\u002Fneuralnetworksanddeeplearning.com\u002F>`_]\n\n* **Deep Learning with Python**:\n  [`Link \u003Chttps:\u002F\u002Fwww.amazon.com\u002FDeep-Learning-Python-Francois-Chollet\u002Fdp\u002F1617294438\u002Fref=as_li_ss_tl?s=books&ie=UTF8&qid=1519989624&sr=1-4&keywords=deep+learning+with+python&linkCode=sl1&tag=trndingcom-20&linkId=ec7663329fdb7ace60f39c762e999683>`_]\n\n* **Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems**:\n  [`Link \u003Chttps:\u002F\u002Fwww.amazon.com\u002FHands-Machine-Learning-Scikit-Learn-TensorFlow\u002Fdp\u002F1491962291\u002Fref=as_li_ss_tl?ie=UTF8&qid=1519989725&sr=1-2-ent&linkCode=sl1&tag=trndingcom-20&linkId=71938c9398940c7b0a811dc1cfef7cc3>`_]\n\n\n************\nBlogs\n************\n\n.. image:: _img\u002Fmainpage\u002FBlogger_icon.png\n\n* **Colah's blog**:\n  [`Link \u003Chttp:\u002F\u002Fcolah.github.io\u002F>`_]\n\n* **Andrej Karpathy blog**:\n  [`Link \u003Chttp:\u002F\u002Fkarpathy.github.io\u002F>`_]\n\n* **The Spectator** Shakir's Machine Learning Blog:\n  [`Link \u003Chttp:\u002F\u002Fblog.shakirm.com\u002F>`_]\n\n* **WILDML**:\n  [`Link \u003Chttp:\u002F\u002Fwww.wildml.com\u002Fabout\u002F>`_]\n\n* **Distill blog** It is more like a journal than a blog because it has a peer review process and only accepted articles will be published on that.:\n  [`Link \u003Chttps:\u002F\u002Fdistill.pub\u002F>`_]\n\n* **BAIR** Berkeley Artificial Inteliigent Research:\n  [`Link \u003Chttp:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002F>`_]\n\n* **Sebastian Ruder's blog**:\n  [`Link \u003Chttp:\u002F\u002Fruder.io\u002F>`_]\n\n* **inFERENCe**:\n  [`Link \u003Chttps:\u002F\u002Fwww.inference.vc\u002Fpage\u002F2\u002F>`_]\n\n* **i am trask** A Machine Learning Craftsmanship Blog:\n  [`Link \u003Chttp:\u002F\u002Fiamtrask.github.io>`_]\n\n\n************\nTutorials\n************\n\n.. image:: _img\u002Fmainpage\u002Ftutorial.png\n\n* **Deep Learning Tutorials**:\n  [`Link \u003Chttp:\u002F\u002Fdeeplearning.net\u002Ftutorial\u002F>`_]\n\n* **Deep Learning for NLP with Pytorch** by Pytorch:\n  [`Link \u003Chttps:\u002F\u002Fpytorch.org\u002Ftutorials\u002Fbeginner\u002Fdeep_learning_nlp_tutorial.html>`_]\n\n* **Deep Learning for Natural Language Processing: Tutorials with Jupyter Notebooks** by Jon Krohn:\n  [`Link \u003Chttps:\u002F\u002Finsights.untapt.com\u002Fdeep-learning-for-natural-language-processing-tutorials-with-jupyter-notebooks-ad67f336ce3f>`_]\n\n\n************\nFrameworks\n************\n\n* **Tensorflow**:\n  [`Link \u003Chttps:\u002F\u002Fwww.tensorflow.org\u002F>`_]\n\n* **Pytorch**:\n  [`Link \u003Chttps:\u002F\u002Fpytorch.org\u002F>`_]\n\n* **CNTK**:\n  [`Link \u003Chttps:\u002F\u002Fdocs.microsoft.com\u002Fen-us\u002Fcognitive-toolkit\u002F>`_]\n\n* **MatConvNet**:\n  [`Link \u003Chttp:\u002F\u002Fwww.vlfeat.org\u002Fmatconvnet\u002F>`_]\n\n* **Keras**:\n  [`Link \u003Chttps:\u002F\u002Fkeras.io\u002F>`_]\n\n* **Caffe**:\n  [`Link \u003Chttp:\u002F\u002Fcaffe.berkeleyvision.org\u002F>`_]\n\n* **Theano**:\n  [`Link \u003Chttp:\u002F\u002Fwww.deeplearning.net\u002Fsoftware\u002Ftheano\u002F>`_]\n\n* **CuDNN**:\n  [`Link \u003Chttps:\u002F\u002Fdeveloper.nvidia.com\u002Fcudnn>`_]\n\n* **Torch**:\n  [`Link \u003Chttps:\u002F\u002Fgithub.com\u002Ftorch\u002Ftorch7>`_]\n\n* **Deeplearning4j**:\n  [`Link \u003Chttps:\u002F\u002Fdeeplearning4j.org\u002F>`_]\n\n\n************\nContributing\n************\n\n\n*For typos, unless significant changes, please do not create a pull request. Instead, declare them in issues or email the repository owner*. Please note we have a code of conduct, please follow it in all your interactions with the project.\n\n========================\nPull Request Process\n========================\n\nPlease consider the following criterions in order to help us in a better way:\n\n1. The pull request is mainly expected to be a link suggestion.\n2. Please make sure your suggested resources are not obsolete or broken.\n3. Ensure any install or build dependencies are removed before the end of the layer when doing a\n   build and creating a pull request.\n4. Add comments with details of changes to the interface, this includes new environment\n   variables, exposed ports, useful file locations and container parameters.\n5. You may merge the Pull Request in once you have the sign-off of at least one other developer, or if you\n   do not have permission to do that, you may request the owner to merge it for you if you believe all checks are passed.\n\n========================\nFinal Note\n========================\n\nWe are looking forward to your kind feedback. Please help us to improve this open source project and make our work better.\nFor contribution, please create a pull request and we will investigate it promptly. Once again, we appreciate\nyour kind feedback and support.\n","###################################################\n深度学习——你需要知道的一切\n###################################################\n\n.. image:: https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fcontributions-welcome-brightgreen.svg?style=flat\n    :target: https:\u002F\u002Fgithub.com\u002Fosforscience\u002Fdeep-learning-all-you-need\u002Fpulls\n.. image:: https:\u002F\u002Fbadges.frapsoft.com\u002Fos\u002Fv2\u002Fopen-source.png?v=103\n    :target: https:\u002F\u002Fgithub.com\u002Fellerbrock\u002Fopen-source-badge\u002F\n.. image:: https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fl\u002Fansicolortags.svg\n      :target: https:\u002F\u002Fgithub.com\u002Fosforscience\u002Fdeep-learning-all-you-need\u002Fblob\u002Fmaster\u002FLICENSE\n.. image:: https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Fmachinemindset.svg?label=Follow&style=social\n      :target: https:\u002F\u002Ftwitter.com\u002Fmachinemindset\n      \n##########################################################################\n赞助\n##########################################################################\n\n为了支持本项目的维护与升级，请您考虑`赞助项目开发者 \u003Chttps:\u002F\u002Fgithub.com\u002Fsponsors\u002Fastorfi\u002Fdashboard>`_。\n\n任何级别的支持都是对本项目的重要贡献 :heart:\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsponsors\u002Fastorfi\u002Fdashboard\" target=\"_blank\">\n  \u003Cimg width=\"600\" height=\"500\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_f47972565d7a.jpg\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>      \n      \n\n###################################################\n免费下载Python机器学习书籍\n###################################################\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"http:\u002F\u002Fwww.machinelearningmindset.com\u002Fdeep-learning-roadmap\u002F\" target=\"_blank\">\n  \u003Cimg width=\"900\" height=\"625\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_acaf49a52074.png\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>\n   \n   \n###################################################\nSlack社区\n###################################################\n\n.. raw:: html\n\n   \u003Cdiv align=\"center\">\n\n.. raw:: html\n\n \u003Ca href=\"https:\u002F\u002Fwww.machinelearningmindset.com\u002Fslack-group\u002F\" target=\"_blank\">\n  \u003Cimg width=\"1033\" height=\"350\" align=\"center\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_readme_2f07fac72f26.png\"\u002F>\n \u003C\u002Fa>\n\n.. raw:: html\n\n   \u003C\u002Fdiv>\n\n\n##################\n目录\n##################\n.. contents::\n  :local:\n  :depth: 4\n\n.. image:: _img\u002Fmainpage\u002Flogo.gif\n\n***************\n引言\n***************\n\n本项目的目的是为开发者和研究人员提供一条捷径，帮助他们快速找到有关深度学习的有用资源。\n\n============\n动机\n============\n\n这个开源项目有多种动机。\n\n.. --------------------\n.. 为什么选择深度学习？\n.. --------------------\n\n------------------------------------------------------------\n这个开源项目的意义何在？\n------------------------------------------------------------\n\n目前已有许多类似本仓库的资源库，它们内容全面且非常实用。说实话，这些资源库让我一度怀疑是否还有必要创建这样一个仓库！\n\n**本仓库的独特之处在于其资源的精准定位**。资源的组织方式使得用户能够轻松找到自己所需的内容。我们把资源分成了众多类别，刚开始可能会让人感到有些复杂！然而，一旦明确了目标，就能迅速找到最相关的资源。即使一开始不知道该寻找什么，也可以先从通用资源入手。\n\n\n.. ================================================\n.. 如何充分利用这一努力\n.. ================================================\n\n************\n论文\n************\n\n.. image:: _img\u002Fmainpage\u002Farticle.jpeg\n\n本章收录了深度学习领域发表的相关论文。\n\n====================\n模型\n====================\n\n-----------------------\n卷积神经网络\n-----------------------\n\n  .. image:: _img\u002Fmainpage\u002Fconvolutional.png\n\n.. 对于连续的线条，必须从同一位置开始。\n* **使用深度卷积神经网络进行ImageNet分类** :\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4824-imagenet-classification-with-deep-convolutional-neural-networks>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fdontfollowmeimcrazy\u002Fimagenet>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **用于句子分类的卷积神经网络** :\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1408.5882>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyoonkim\u002FCNN_sentence>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **大规模视频分类中的卷积神经网络** :\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FKarpathy_Large-scale_Video_Classification_2014_CVPR_paper.html>`_][`项目页面 \u003Chttps:\u002F\u002Fcs.stanford.edu\u002Fpeople\u002Fkarpathy\u002Fdeepvideo\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **利用卷积神经网络学习并迁移图像中层特征表示** :\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FOquab_Learning_and_Transferring_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n\n* **深度卷积神经网络在LVCSR中的应用** :\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6639347\u002F&hl=zh-CN&sa=T&oi=gsb&ct=res&cd=0&ei=KknXWYbGFMbFjwSsyICADQ&scisig=AAGBfm2F0Zlu0ciUwadzshNNm80IQQhuhA>`_]\n  \n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **人脸识别：基于卷积神经网络的方法** :\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F554195\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n\n-----------------------\n循环神经网络\n-----------------------\n\n  .. image:: _img\u002Fmainpage\u002FRecurrent_neural_network_unfold.svg\n\n\n.. 对于连续的线条，必须从同一位置开始。\n* **循环神经网络架构的实证探索** :\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fjozefowicz15.pdf?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=revue>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fdebajyotidatta\u002FRecurrentArchitectures>`_]\n\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **LSTM：搜索空间之旅** :\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F7508408\u002F>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Ffomorians\u002Flstm-odyssey>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n* **训练循环神经网络的困难性** :\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv28\u002Fpascanu13.pdf>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fpascanur\u002FtrainingRNNs>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **学会遗忘：使用LSTM进行持续预测** :\n  [`论文 \u003Chttp:\u002F\u002Fdigital-library.theiet.org\u002Fcontent\u002Fconferences\u002F10.1049\u002Fcp_19991218>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n-----------------------\n自编码器\n-----------------------\n\n.. image:: _img\u002Fmainpage\u002FAutoencoder_structure.png\n\n* **使用去噪自编码器提取和组合鲁棒特征**：\n  [`论文 \u003Chttps:\u002F\u002Fdl.acm.org\u002Fcitation.cfm?id=1390294>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **堆叠式去噪自编码器：通过局部去噪准则在深度网络中学习有用表示**：\n  [`论文 \u003Chttp:\u002F\u002Fwww.jmlr.org\u002Fpapers\u002Fv11\u002Fvincent10a.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Frajarsheem\u002Flibsdae-autoencoder-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **对抗自编码器**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.05644>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fconan7882\u002Fadversarial-autoencoders>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **自编码器、无监督学习与深度架构**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv27\u002Fbaldi12a\u002Fbaldi12a.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **利用神经网络降低数据维度**：\n  [`论文 \u003Chttp:\u002F\u002Fscience.sciencemag.org\u002Fcontent\u002F313\u002F5786\u002F504>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fjordn\u002Fautoencoder>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n-----------------------\n生成模型\n-----------------------\n\n.. image:: _img\u002Fmainpage\u002Fgenerative.png\n\n* **利用生成模型进行判别分类**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F1520-exploiting-generative-models-in-discriminative-classifiers.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **基于深度生成模型的半监督学习**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5352-semi-supervised-learning-with-deep-generative-models>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fwohlert\u002Fsemi-supervised-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n* **生成对抗网络**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5423-generative-adversarial-nets>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fgoodfeli\u002Fadversarial>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **广义去噪自编码器作为生成模型**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5023-generalized-denoising-auto-encoders-as-generative-models>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **基于深度卷积生成对抗网络的无监督表征学习**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06434>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002FDCGAN-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n-----------------------\n概率模型\n-----------------------\n\n* **深度生成模型中的随机反向传播与近似推断**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1401.4082>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **认知的概率模型：探索表征与归纳偏置**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1364661310001129>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **关于深度生成模型及其在识别中的应用**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F5995710\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n\n\n\n====================\n核心\n====================\n\n---------------------\n优化\n---------------------\n\n.. ################################################################################\n.. 对于连续的行，每一行必须从相同的位置开始。\n* **批量归一化：通过减少内部协变量偏移加速深度网络训练**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1502.03167>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **Dropout：防止神经网络过拟合的简单方法**：\n  [`论文 \u003Chttp:\u002F\u002Fwww.jmlr.org\u002Fpapers\u002Fvolume15\u002Fsrivastava14a\u002Fsrivastava14a.pdf?utm_content=buffer79b43&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **训练非常深的网络**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5850-training-very-deep-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **深入研究ReLU激活函数：在ImageNet分类任务上超越人类水平性能**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FHe_Delving_Deep_into_ICCV_2015_paper.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **大规模分布式深度网络**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4687-large-scale-distributed-deep-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n------------------------\n表征学习\n------------------------\n\n* **基于深度卷积生成对抗网络的无监督表征学习**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06434>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002FNewmu\u002Fdcgan_code>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **表征学习：综述与新视角**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6472238\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **InfoGAN：通过信息最大化生成对抗网络实现可解释的表征学习**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F6399-infogan-interpretable-representation>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fopenai\u002FInfoGAN>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n------------------------------------\n理解与迁移学习\n------------------------------------\n\n* **利用卷积神经网络学习并迁移图像中层表征**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FOquab_Learning_and_Transferring_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **知识蒸馏**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1503.02531>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **DeCAF：用于通用视觉识别的深度卷积激活特征**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv32\u002Fdonahue14.pdf>`_][\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **深度神经网络中的特征有多大的可迁移性？**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5347-how-transferable-are-features-in-deep-n%E2%80%A6>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyosinski\u002Fconvnet_transfer>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n-----------------------\n强化学习\n-----------------------\n\n* **通过深度强化学习实现人类水平控制**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.nature.com\u002Farticles\u002Fnature14236\u002F>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fdevsisters\u002FDQN-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **用深度强化学习玩Atari游戏**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.5602>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002Fdeep-rl-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **深度强化学习中的连续控制**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1509.02971>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fstevenpjg\u002Fddpg-aigym>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **采用双重Q学习的深度强化学习**：\n  [`论文 \u003Chttp:\u002F\u002Fwww.aaai.org\u002Focs\u002Findex.php\u002FAAAI\u002FAAAI16\u002Fpaper\u002Fdownload\u002F12389\u002F11847>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fcarpedm20\u002Fdeep-rl-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **深度强化学习中的对决网络架构**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1511.06581>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyoosan\u002Fdeeprl>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n\n====================\n应用\n====================\n\n--------------------\n图像识别\n--------------------\n\n* **用于图像识别的深度残差学习**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fhtml\u002FHe_Deep_Residual_Learning_CVPR_2016_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fgcr\u002Ftorch-residual-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **用于大规模图像识别的非常深的卷积神经网络**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1409.1556>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **用于图像分类的多列深度神经网络**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1202.2745>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **DeepID3：使用非常深的神经网络进行人脸识别**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1502.00873>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **深入卷积神经网络：可视化图像分类模型与显著性图**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.6034>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fartvandelay\u002FDeep_Inside_Convolutional_Networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Deep Image：扩展图像识别规模**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fvc\u002Farxiv\u002Fpapers\u002F1501\u002F1501.02876v1.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **用于视觉识别与描述的长期循环卷积网络**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FDonahue_Long-Term_Recurrent_Convolutional_2015_CVPR_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002FJaggerYoung\u002FLRCN-for-Activity-Recognition>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **用于跨音频-视觉匹配识别的3D卷积神经网络**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F8063416>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fastorfi\u002Flip-reading-deeplearning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n--------------------\n目标识别\n--------------------\n\n* **使用深度卷积神经网络进行ImageNet分类**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4824-imagenet-classification-with-deep-convolutional-neural-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **利用Places数据库学习场景识别的深度特征**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5349-learning-deep-features>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **使用深度神经网络进行可扩展的目标检测**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fhtml\u002FErhan_Scalable_Object_Detection_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **Faster R-CNN：通过区域提议网络实现实时目标检测**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5638-faster-r-cnn-towards-real-time-object-detection-with-region-proposal-networks>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Frbgirshick\u002Fpy-faster-rcnn>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **OverFeat：使用卷积网络实现集成识别、定位与检测**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1312.6229>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fsermanet\u002FOverFeat>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **现成的CNN特征：令人惊叹的识别基线**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_workshops_2014\u002FW15\u002Fhtml\u002FRazavian_CNN_Features_Off-the-Shelf_2014_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **用于目标识别的最佳多阶段架构是什么？**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F5459469\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_2.png\n\n\n--------------------\n动作识别\n--------------------\n\n* **用于视觉识别与描述的长期循环卷积网络**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FDonahue_Long-Term_Recurrent_Convolutional_2015_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **使用3D卷积网络学习时空特征**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fhtml\u002FTran_Learning_Spatiotemporal_Features_ICCV_2015_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002FDavideA\u002Fc3d-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **通过利用时间结构描述视频**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fhtml\u002FYao_Describing_Videos_by_ICCV_2015_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Ftsenghungchen\u002FSA-tensorflow>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **用于视频动作识别的卷积双流网络融合**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fhtml\u002FFeichtenhofer_Convolutional_Two-Stream_Network_CVPR_2016_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Ffeichtenhofer\u002Ftwostreamfusion>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **时间片段网络：迈向深度动作识别的良好实践**：\n  [`论文 \u003Chttps:\u002F\u002Flink.springer.com\u002Fchapter\u002F10.1007\u002F978-3-319-46484-8_2>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyjxiong\u002Ftemporal-segment-networks>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n----------------------------\n标题生成\n----------------------------\n\n* **展示、注意与讲述：带有视觉注意力的神经图像标题生成**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fxuc15.pdf>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyunjey\u002Fshow-attend-and-tell>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **心灵之眼：用于图像标题生成的递归视觉表示**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FChen_Minds_Eye_A_2015_CVPR_paper.html>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_2.png\n\n* **生成对抗文本到图像合成**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Freed16.pdf>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fzsdonghao\u002Ftext-to-image>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **用于生成图像描述的深度视觉-语义对齐**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FKarpathy_Deep_Visual-Semantic_Alignments_2015_CVPR_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fjonkuo\u002FDeep-Learning-Image-Captioning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **展示与讲述：一种神经图像标题生成器**：\n  [`论文 \u003Chttps:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fhtml\u002FVinyals_Show_and_Tell_2015_CVPR_paper.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002FDeepRNN\u002Fimage_captioning>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n\n----------------------------\n自然语言处理\n----------------------------\n\n* **词和短语的分布式表示及其组合性**：\n  [`论文 \u003Chttp:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf>`_][`代码 \u003Chttps:\u002F\u002Fcode.google.com\u002Farchive\u002Fp\u002Fword2vec\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **向量空间中词表示的有效估计**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1301.3781.pdf>`_][`代码 \u003Chttps:\u002F\u002Fcode.google.com\u002Farchive\u002Fp\u002Fword2vec\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **基于神经网络的序列到序列学习**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1409.3215.pdf>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Ffarizrahman4u\u002Fseq2seq>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **通过联合学习对齐与翻译实现神经机器翻译**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fpdf\u002F1409.0473.pdf>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fnmt>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **直奔主题：使用指针生成器网络进行摘要生成**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1704.04368>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fabisee\u002Fpointer-generator>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **注意力就是一切**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fjadore801120\u002Fattention-is-all-you-need-pytorch>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **用于句子分类的卷积神经网络**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1408.5882>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fyoonkim\u002FCNN_sentence>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n----------------------------\n语音技术\n----------------------------\n\n* **深度神经网络在语音识别声学建模中的应用：四个研究小组的共同见解**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6296526\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_5.png\n\n* **迈向端到端的循环神经网络语音识别**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv32\u002Fgraves14.pdf>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **基于深度循环神经网络的语音识别**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6638947\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **快速且准确的循环神经网络声学模型用于语音识别**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1507.06947>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **Deep Speech 2：英语和普通话的端到端语音识别**：\n  [`论文 \u003Chttp:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Famodei16.html>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002FDeepSpeech>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n* **一种新颖的基于音素感知深度神经网络的说话人识别方案**：\n  [`论文 \u003Chttps:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F6853887\u002F>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_3.png\n\n* **基于3D卷积神经网络的文本无关说话人验证**：\n  [`论文 \u003Chttps:\u002F\u002Farxiv.org\u002Fabs\u002F1705.09422>`_][`代码 \u003Chttps:\u002F\u002Fgithub.com\u002Fastorfi\u002F3D-convolutional-speaker-recognition>`_]\n\n  .. image:: _img\u002Fmainpage\u002Fstar_4.png\n\n\n************\n数据集\n************\n\n====================\n图像\n====================\n\n\n----------------------------\n通用\n----------------------------\n\n* **MNIST** 手写数字数据集：\n  [`链接 \u003Chttp:\u002F\u002Fyann.lecun.com\u002Fexdb\u002Fmnist\u002F>`_]\n\n\n----------------------------\n人脸\n----------------------------\n\n* **人脸识别技术（FERET）** FERET计划的目标是开发自动人脸识别能力，以协助安保、情报和执法部门人员履行其职责：\n  [`链接 \u003Chttps:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-technology-feret>`_]\n\n* **卡内基梅隆大学姿态、光照和表情（PIE）人脸数据库** 在2000年10月至12月期间，我们收集了68位人士的41,368张图像：\n  [`链接 \u003Chttps:\u002F\u002Fwww.ri.cmu.edu\u002Fpublications\u002Fthe-cmu-pose-illumination-and-expression-pie-database-of-human-faces\u002F>`_]\n\n* **YouTube Faces DB** 该数据集包含来自1595个人的3425个视频。所有视频均从YouTube下载。每个对象平均有2.15个视频可用：\n  [`链接 \u003Chttps:\u002F\u002Fwww.cs.tau.ac.il\u002F~wolf\u002Fytfaces\u002F>`_]\n\n* **语法面部表情数据集** 为辅助面部表情的自动化分析而开发：\n  [`链接 \u003Chttps:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FGrammatical+Facial+Expressions>`_]\n\n* **FaceScrub** 包含超过10万张530个人脸图像的数据集：\n  [`链接 \u003Chttp:\u002F\u002Fvintage.winklerbros.net\u002Ffacescrub.html>`_]\n\n* **IMDB-WIKI** 包含50多万张带年龄和性别标签的人脸图像：\n  [`链接 \u003Chttps:\u002F\u002Fdata.vision.ee.ethz.ch\u002Fcvl\u002Frrothe\u002Fimdb-wiki\u002F>`_]\n\n* **FDDB** 面部检测数据集和基准测试（FDDB）：\n  [`链接 \u003Chttp:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002F>`_]\n\n----------------------------\n物体识别\n----------------------------\n\n* **COCO** Microsoft COCO：上下文中的常见物体：\n  [`链接 \u003Chttp:\u002F\u002Fcocodataset.org\u002F#home>`_]\n\n* **ImageNet** 著名的ImageNet数据集：\n  [`链接 \u003Chttp:\u002F\u002Fwww.image-net.org\u002F>`_]\n\n* **Open Images Dataset** Open Images是一个包含约900万张图片的数据集，这些图片已被标注了图像级别标签和物体边界框：\n  [`链接 \u003Chttps:\u002F\u002Fstorage.googleapis.com\u002Fopenimages\u002Fweb\u002Findex.html>`_]\n\n* **Caltech-256物体类别数据集** 一个大型物体分类数据集：\n  [`链接 \u003Chttps:\u002F\u002Fauthors.library.caltech.edu\u002F7694\u002F>`_]\n\n* **Pascal VOC数据集** 一个用于分类任务的大规模数据集：\n  [`链接 \u003Chttp:\u002F\u002Fhost.robots.ox.ac.uk\u002Fpascal\u002FVOC\u002F>`_]\n\n* **CIFAR 10 \u002F CIFAR 100** CIFAR-10数据集由10个类别的60000张32x32彩色图像组成。CIFAR-100与CIFAR-10类似，但它有100个类，每个类包含600张图像：\n  [`链接 \u003Chttps:\u002F\u002Fwww.cs.toronto.edu\u002F~kriz\u002Fcifar.html>`_]\n\n\n----------------------------\n动作识别\n----------------------------\n\n* **HMDB** 一个大型人体动作数据库：\n  [`链接 \u003Chttp:\u002F\u002Fserre-lab.clps.brown.edu\u002Fresource\u002Fhmdb-a-large-human-motion-database\u002F>`_]\n\n* **MHAD** 伯克利多模态人类动作数据库：\n  [`链接 \u003Chttp:\u002F\u002Ftele-immersion.citris-uc.org\u002Fberkeley_mhad>`_]\n\n* **UCF101 - 动作识别数据集** UCF101是一个包含101个动作类别的真实动作视频的动作识别数据集，这些视频均来自YouTube。该数据集是UCF50数据集的扩展，后者只有50个动作类别：\n  [`链接 \u003Chttp:\u002F\u002Fcrcv.ucf.edu\u002Fdata\u002FUCF101.php>`_]\n\n* **THUMOS数据集** 一个用于动作分类的大规模数据集：\n  [`链接 \u003Chttp:\u002F\u002Fcrcv.ucf.edu\u002Fdata\u002FTHUMOS.php>`_]\n\n* **ActivityNet** 一个用于理解人类活动的大规模视频基准测试：\n  [`链接 \u003Chttp:\u002F\u002Factivity-net.org\u002F>`_]\n\n======================================\n文本与自然语言处理\n======================================\n\n\n-----------------------\n通用\n-----------------------\n\n* **10亿词语言模型基准**：该项目旨在为语言建模实验提供标准化的训练和测试设置：\n  [`链接 \u003Chttp:\u002F\u002Fwww.statmt.org\u002Flm-benchmark\u002F>`_]\n\n* **Common Crawl**：Common Crawl语料库包含过去7年间收集的数PB数据，其中包括原始网页数据、提取的元数据以及文本内容：\n  [`链接 \u003Chttp:\u002F\u002Fcommoncrawl.org\u002Fthe-data\u002Fget-started\u002F>`_]\n\n* **Yelp开放数据集**：Yelp的企业、评论和用户数据的一个子集，可用于个人、教育和学术目的：\n  [`链接 \u003Chttps:\u002F\u002Fwww.yelp.com\u002Fdataset>`_]\n\n\n-----------------------\n文本分类\n-----------------------\n\n* **20 Newsgroups**：20 Newsgroups数据集包含约2万篇新闻组文档，几乎平均分布在20个不同的新闻组中：\n  [`链接 \u003Chttp:\u002F\u002Fqwone.com\u002F~jason\u002F20Newsgroups\u002F>`_]\n\n* **广播新闻**：1996年广播新闻语音语料库共包含来自ABC、CNN和CSPAN电视台以及NPR和PRI电台的104小时广播节目，并附有相应的文字稿：\n  [`链接 \u003Chttps:\u002F\u002Fcatalog.ldc.upenn.edu\u002FLDC97S44>`_]\n\n* **Wikitext长期依赖语言模型数据集**：该数据集由维基百科上经过验证的好条目和特色条目中提取的超过1亿个词素组成：\n  [`链接 \u003Chttps:\u002F\u002Feinstein.ai\u002Fresearch\u002Fthe-wikitext-long-term-dependency-language-modeling-dataset>`_]\n\n-----------------------\n问答\n-----------------------\n\n* **Deep Mind与牛津大学联合发布的问答语料库**，包含来自CNN和Daily Mail网站的大约百万篇新闻文章及其相关问题的两个新语料库：\n  [`链接 \u003Chttps:\u002F\u002Fgithub.com\u002Fdeepmind\u002Frc-data>`_]\n\n* **斯坦福问答数据集（SQuAD）**：该数据集由众包工作者针对维基百科文章提出的问题组成：\n  [`链接 \u003Chttps:\u002F\u002Frajpurkar.github.io\u002FSQuAD-explorer\u002F>`_]\n\n* **亚马逊问答数据集**：包含来自亚马逊的问答数据，共计约140万个已回答的问题：\n  [`链接 \u003Chttp:\u002F\u002Fjmcauley.ucsd.edu\u002Fdata\u002Famazon\u002Fqa\u002F>`_]\n\n\n\n-----------------------\n情感分析\n-----------------------\n\n* **多领域情感数据集**：该数据集包含了从Amazon.com上获取的多种产品类别的商品评论：\n  [`链接 \u003Chttp:\u002F\u002Fwww.cs.jhu.edu\u002F~mdredze\u002Fdatasets\u002Fsentiment\u002F>`_]\n\n* **斯坦福情感树库数据集**：斯坦福情感树库是首个带有完整标注句法树的语料库，能够对语言中情感成分的组合效应进行全面分析：\n  [`链接 \u003Chttps:\u002F\u002Fnlp.stanford.edu\u002Fsentiment\u002F>`_]\n\n* **大型电影评论数据集**：这是一个用于二元情感分类的数据集：\n  [`链接 \u003Chttp:\u002F\u002Fai.stanford.edu\u002F~amaas\u002Fdata\u002Fsentiment\u002F>`_]\n\n\n-----------------------\n机器翻译\n-----------------------\n\n* **加拿大第36届议会齐次汉萨德语料库**：该语料库包含130万对齐的文本片段：\n  [`链接 \u003Chttps:\u002F\u002Fwww.isi.edu\u002Fnatural-language\u002Fdownload\u002Fhansard\u002F>`_]\n\n* **Europarl：统计机器翻译平行语料库**：该语料库取自欧洲议会会议记录：\n  [`链接 \u003Chttp:\u002F\u002Fwww.statmt.org\u002Feuroparl\u002F>`_]\n\n\n-----------------------\n摘要生成\n-----------------------\n\n* **法律案例报告数据集**：这是一个包含4000个法律案例的文本语料库，可用于自动摘要和引文分析：\n  [`链接 \u003Chttps:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FLegal+Case+Reports>`_]\n\n\n======================================\n语音技术\n======================================\n\n* **TIMIT连续语音声学-音系学语料库**：TIMIT朗读语音语料库旨在为声学-音系学研究以及自动语音识别系统的开发和评估提供语音数据：\n  [`链接 \u003Chttps:\u002F\u002Fcatalog.ldc.upenn.edu\u002Fldc93s1>`_]\n\n* **LibriSpeech**：LibriSpeech是一个约1000小时的16kHz英语朗读语音语料库，由Vassil Panayotov在Daniel Povey的协助下整理而成：\n  [`链接 \u003Chttp:\u002F\u002Fwww.openslr.org\u002F12\u002F>`_]\n\n* **VoxCeleb**：一个大规模的视听数据集：\n  [`链接 \u003Chttp:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvoxceleb\u002F>`_]\n\n* **NIST说话人识别**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fiad\u002Fmig\u002Fspeaker-recognition>`_]\n\n\n\n\n\n\n************\n课程\n************\n\n.. image:: _img\u002Fmainpage\u002Fonline.png\n\n* **斯坦福大学Coursera平台上的机器学习课程**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fmachine-learning>`_]\n\n* **Coursera平台上的神经网络与深度学习专项课程**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fneural-networks-deep-learning>`_]\n\n* **Google推出的深度学习入门课程**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fdeep-learning--ud730>`_]\n\n* **卡内基梅隆大学的深度学习导论课程**：\n  [`链接 \u003Chttp:\u002F\u002Fdeeplearning.cs.cmu.edu\u002F>`_]\n\n* **NVIDIA深度学习学院提供的课程**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.nvidia.com\u002Fen-us\u002Fdeep-learning-ai\u002Feducation\u002F>`_]\n\n* **斯坦福大学的视觉识别卷积神经网络课程**：\n  [`链接 \u003Chttp:\u002F\u002Fcs231n.stanford.edu\u002F>`_]\n\n* **斯坦福大学自然语言处理深度学习课程**：\n  [`链接 \u003Chttp:\u002F\u002Fcs224d.stanford.edu\u002F>`_]\n\n* **fast.ai提供的深度学习课程**：\n  [`链接 \u003Chttp:\u002F\u002Fwww.fast.ai\u002F>`_]\n\n* **印度理工学院Kharagpur分校关于视觉计算深度学习的课程**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLuv3GM6-gsE1Biyakccxb3FAn4wBLyfWf>`_]\n\n\n\n************\n书籍\n************\n\n.. image:: _img\u002Fmainpage\u002Fbooks.jpg\n\n* **Ian Goodfellow著《深度学习》**：\n  [`链接 \u003Chttp:\u002F\u002Fwww.deeplearningbook.org\u002F>`_]\n\n* **《神经网络与深度学习》**：\n  [`链接 \u003Chttp:\u002F\u002Fneuralnetworksanddeeplearning.com\u002F>`_]\n\n* **《用Python进行深度学习》**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.amazon.com\u002FDeep-Learning-Python-Francois-Chollet\u002Fdp\u002F1617294438\u002Fref=as_li_ss_tl?s=books&ie=UTF8&qid=1519989624&sr=1-4&keywords=deep+learning+with+python&linkCode=sl1&tag=trndingcom-20&linkId=ec7663329fdb7ace60f39c762e999683>`_]\n\n* **《动手学机器学习：使用Scikit-Learn和TensorFlow构建智能系统》**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.amazon.com\u002FHands-Machine-Learning-Scikit-Learn-TensorFlow\u002Fdp\u002F1491962291\u002Fref=as_li_ss_tl?ie=UTF8&qid=1519989725&sr=1-2-ent&linkCode=sl1&tag=trndingcom-20&linkId=71938c9398940c7b0a811dc1cfef7cc3>`_]\n\n\n************\n博客\n************\n\n.. image:: _img\u002Fmainpage\u002FBlogger_icon.png\n\n* **Colah的博客**：\n  [`链接 \u003Chttp:\u002F\u002Fcolah.github.io\u002F>`_]\n\n* **Andrej Karpathy的博客**：\n  [`链接 \u003Chttp:\u002F\u002Fkarpathy.github.io\u002F>`_]\n\n* **The Spectator Shakir的机器学习博客**：\n  [`链接 \u003Chttp:\u002F\u002Fblog.shakirm.com\u002F>`_]\n\n* **WILDML**：\n  [`链接 \u003Chttp:\u002F\u002Fwww.wildml.com\u002Fabout\u002F>`_]\n\n* **Distill博客**：它更像一本期刊而非博客，因为其采用同行评审机制，只有通过评审的文章才会被发表：\n  [`链接 \u003Chttps:\u002F\u002Fdistill.pub\u002F>`_]\n\n* **BAIR伯克利人工智能研究中心**：\n  [`链接 \u003Chttp:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002F>`_]\n\n* **塞巴斯蒂安·鲁德尔的博客**：\n  [`链接 \u003Chttp:\u002F\u002Fruder.io\u002F>`_]\n\n* **inFERENCe**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.inference.vc\u002Fpage\u002F2\u002F>`_]\n\n* **i am trask 机器学习工艺博客**：\n  [`链接 \u003Chttp:\u002F\u002Fiamtrask.github.io>`_]\n\n\n************\n教程\n************\n\n.. image:: _img\u002Fmainpage\u002Ftutorial.png\n\n* **深度学习教程**：\n  [`链接 \u003Chttp:\u002F\u002Fdeeplearning.net\u002Ftutorial\u002F>`_]\n\n* **PyTorch 官方的 NLP 深度学习教程**：\n  [`链接 \u003Chttps:\u002F\u002Fpytorch.org\u002Ftutorials\u002Fbeginner\u002Fdeep_learning_nlp_tutorial.html>`_]\n\n* **乔恩·克罗恩的自然语言处理深度学习教程（附 Jupyter 笔记本）**：\n  [`链接 \u003Chttps:\u002F\u002Finsights.untapt.com\u002Fdeep-learning-for-natural-language-processing-tutorials-with-jupyter-notebooks-ad67f336ce3f>`_]\n\n\n************\n框架\n************\n\n* **TensorFlow**：\n  [`链接 \u003Chttps:\u002F\u002Fwww.tensorflow.org\u002F>`_]\n\n* **PyTorch**：\n  [`链接 \u003Chttps:\u002F\u002Fpytorch.org\u002F>`_]\n\n* **CNTK**：\n  [`链接 \u003Chttps:\u002F\u002Fdocs.microsoft.com\u002Fen-us\u002Fcognitive-toolkit\u002F>`_]\n\n* **MatConvNet**：\n  [`链接 \u003Chttp:\u002F\u002Fwww.vlfeat.org\u002Fmatconvnet\u002F>`_]\n\n* **Keras**：\n  [`链接 \u003Chttps:\u002F\u002Fkeras.io\u002F>`_]\n\n* **Caffe**：\n  [`链接 \u003Chttp:\u002F\u002Fcaffe.berkeleyvision.org\u002F>`_]\n\n* **Theano**：\n  [`链接 \u003Chttp:\u002F\u002Fwww.deeplearning.net\u002Fsoftware\u002Ftheano\u002F>`_]\n\n* **CuDNN**：\n  [`链接 \u003Chttps:\u002F\u002Fdeveloper.nvidia.com\u002Fcudnn>`_]\n\n* **Torch**：\n  [`链接 \u003Chttps:\u002F\u002Fgithub.com\u002Ftorch\u002Ftorch7>`_]\n\n* **Deeplearning4j**：\n  [`链接 \u003Chttps:\u002F\u002Fdeeplearning4j.org\u002F>`_]\n\n\n************\n贡献说明\n************\n\n\n*对于拼写错误，除非是重大修改，请不要提交拉取请求。请在问题中指出或直接发送邮件给仓库负责人*。请注意，我们有一份行为准则，请在与本项目的任何互动中遵守。\n\n========================\n拉取请求流程\n========================\n\n为了帮助我们更好地工作，请您考虑以下几点：\n\n1. 拉取请求主要应为链接建议。\n2. 请确保您建议的资源未过时或失效。\n3. 在构建并提交拉取请求之前，请确保已移除所有安装或构建依赖项。\n4. 添加注释，详细说明接口的更改内容，包括新增的环境变量、开放的端口、重要文件路径以及容器参数。\n5. 当至少有一位其他开发人员签字确认后，您可以合并拉取请求；如果您没有权限执行此操作，且认为所有检查均已通过，可以请求仓库负责人代为合并。\n\n========================\n最后说明\n========================\n\n我们期待您的宝贵反馈。请帮助我们改进这个开源项目，使我们的工作更加出色。如需贡献，请创建一个拉取请求，我们将尽快进行审核。再次感谢您的反馈与支持。","# deep-learning-roadmap 快速上手指南\n\n`deep-learning-roadmap` 并非一个需要安装运行的软件库，而是一个**深度学习资源索引与学习路线图**。它汇集了经典的论文、代码实现、核心概念（如卷积网络、循环网络、生成模型等）以及优化技巧。\n\n本指南将帮助你快速获取该资源库的内容，并指引你如何利用其中的链接开始学习。\n\n## 环境准备\n\n由于本项目主要是文档和资源链接集合，无需复杂的系统依赖。你只需要具备以下基础环境即可浏览和访问相关资源：\n\n*   **操作系统**：Windows \u002F macOS \u002F Linux\n*   **浏览器**：推荐 Chrome 或 Firefox，用于访问 GitHub 页面及论文链接。\n*   **开发环境（可选）**：如果你打算运行资源库中链接的代码示例，建议安装：\n    *   Python 3.8+\n    *   Git (用于克隆仓库)\n    *   深度学习框架：PyTorch 或 TensorFlow (根据具体论文代码需求)\n\n## 获取资源\n\n你可以通过以下两种方式获取该路线图内容：\n\n### 方式一：在线浏览（推荐）\n直接访问项目主页，查看分类清晰的资源目录和免费电子书。\n*   **项目主页**: [https:\u002F\u002Fgithub.com\u002Fmachinelearningmindset\u002Fdeep-learning-roadmap](https:\u002F\u002Fgithub.com\u002Fmachinelearningmindset\u002Fdeep-learning-roadmap)\n*   **免费电子书**: [Download Free Python Machine Learning Book](http:\u002F\u002Fwww.machinelearningmindset.com\u002Fdeep-learning-roadmap\u002F)\n\n### 方式二：本地克隆\n如果你希望离线查阅或贡献内容，可以使用 Git 克隆仓库。\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fmachinelearningmindset\u002Fdeep-learning-roadmap.git\ncd deep-learning-roadmap\n```\n\n> **提示**：如果在国内访问 GitHub 速度较慢，可以考虑使用国内镜像站（如 Gitee 上的镜像，若有）或配置代理加速。\n\n## 基本使用\n\n本项目的使用核心在于**按需检索**。资源已按模型类型和核心主题进行了详细分类。\n\n### 1. 查找特定模型的论文与代码\n假设你想学习 **卷积神经网络 (Convolutional Networks)** 中的经典论文 `Imagenet classification with deep convolutional neural networks`：\n\n1.  打开仓库中的 `README` 文件或在线页面。\n2.  定位到 **Models** -> **Convolutional Networks** 章节。\n3.  点击对应的 [`Paper`](http:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F4824-imagenet-classification-with-deep-convolutional-neural-networks) 链接阅读原文。\n4.  点击对应的 [`Code`](https:\u002F\u002Fgithub.com\u002Fdontfollowmeimcrazy\u002Fimagenet) 链接查看实现代码。\n\n### 2. 学习核心优化技巧\n若想研究 **优化 (Optimization)** 相关的技术（如 Batch Normalization, Dropout）：\n\n1.  导航至 **Core** -> **Optimization** 章节。\n2.  查阅列出的关键论文，例如：\n    *   `Batch Normalization: Accelerating Deep Network Training...`\n    *   `Dropout: A Simple Way to Prevent Neural Networks from Overfitting`\n\n### 3. 加入社区交流\n项目提供了 Slack 群组供开发者交流：\n*   访问 [Slack Group](https:\u002F\u002Fwww.machinelearningmindset.com\u002Fslack-group\u002F) 加入讨论。\n\n---\n**总结**：将 `deep-learning-roadmap` 作为你的深度学习“地图”，遇到不懂的模型或算法时，在此查找对应的经典论文和开源代码实现，是最高效的学习路径。","某初创公司的算法工程师李明接到任务，需在两周内为医疗影像项目搭建一个基于深度学习的原型系统，但他对快速迭代的领域缺乏系统性认知。\n\n### 没有 deep-learning-roadmap 时\n- **资源检索如大海捞针**：在 GitHub 和谷歌上搜索\"Deep Learning resources\"，返回成千上万个结果，难以分辨哪些是过时教程，哪些是核心论文。\n- **学习路径支离破碎**：今天看一篇关于 CNN 的博客，明天找一个 RNN 的视频，知识点零散，无法构建从基础数学到前沿架构的完整知识树。\n- **关键资料遗漏风险高**：由于缺乏权威指引，容易错过该领域必读的经典论文（如 ResNet, Transformer）或最新的 SOTA 实现代码。\n- **时间成本高昂**：花费了整整一周时间在筛选和验证资料真伪上，导致实际编码和模型调优的时间被严重压缩，项目进度告急。\n\n### 使用 deep-learning-roadmap 后\n- **目标资源精准直达**：利用其高度分类的资源目录，李明直接定位到“计算机视觉”和“医学图像分析”板块，瞬间获取经过社区验证的高质量链接。\n- **结构化学习路线清晰**：遵循项目提供的从基础理论到高级应用的路线图，他按部就班地补齐了数学基础，并迅速掌握了当前主流的模型架构。\n- **核心文献一网打尽**：通过\"Papers\"章节，他快速获取了领域内最关键的学术论文列表，确保了技术方案的前沿性和可靠性。\n- **研发效率显著提升**：原本用于搜索资料的一周时间被节省下来，全部投入到数据预处理和模型训练中，提前三天完成了高精度原型的交付。\n\ndeep-learning-roadmap 通过将海量碎片化信息重构为有序的知识地图，帮助开发者从“盲目搜索”转向“高效执行”，极大缩短了从入门到实战的路径。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Finstillai_deep-learning-roadmap_0c5c2d36.png","instillai","Instill AI","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Finstillai_db38df51.png","A company offering AI-based solutions to real-world applications.",null,"contact@instillai.com","https:\u002F\u002Finstillai.com","https:\u002F\u002Fgithub.com\u002Finstillai",[84],{"name":85,"color":86,"percentage":87},"Python","#3572A5",100,4637,664,"2026-04-04T20:12:10","MIT",1,"","未说明",{"notes":96,"python":94,"dependencies":97},"该项目是一个深度学习资源路线图和论文\u002F代码合集，并非一个可直接运行的软件工具或模型库。README 内容主要包含分类整理的学术论文链接、部分项目代码仓库的外部链接以及书籍下载指引。因此，该项目本身没有特定的操作系统、GPU、内存、Python 版本或依赖库的安装需求。用户若需运行其中链接的具体代码示例，需参考各个独立项目仓库的具体要求。",[],[13],[100,101],"deep-learning","resources","2026-03-27T02:49:30.150509","2026-04-06T05:35:48.728379",[],[]]