##机器学习(Machine Learning)&深度学习(Deep Learning)资料(Chapter 2)
—
#####注:机器学习资料[篇目一](https://github.com/ty4z2008/Qix/blob/master/dl.md)共500条,[篇目二](https://github.com/ty4z2008/Qix/blob/master/dl2.md)开始更新
—
—
#####希望转载的朋友**一定要保留原文链接**，因为这个项目还在继续也在不定期更新．希望看到文章的朋友能够学到更多．此外:某些资料在中国访问需要梯子.
—
* [《Image Scaling using Deep Convolutional Neural Networks》](http://engineering.flipboard.com/2015/05/scaling-convnets/)
介绍:使用卷积神经网络的图像缩放.
* [《Proceedings of The 32nd International Conference on Machine Learning》](http://jmlr.org/proceedings/papers/v37/)
介绍:ICML2015 论文集,优化4个+稀疏优化1个；强化学习4个，深度学习3个+深度学习计算1个；贝叶斯非参、高斯过程和学习理论3个；还有计算广告和社会选择.[ICML2015 Sessions](http://icml.cc/2015/?page_id=825).
* [《Image Scaling using Deep Convolutional Neural Networks》](http://engineering.flipboard.com/2015/05/scaling-convnets/)
介绍:使用卷积神经网络的图像缩放.
* [《Microsoft researchers accelerate computer vision accuracy and improve 3D scanning models》](http://blogs.technet.com/b/inside_microsoft_research/archive/2015/06/08/microsoft-researchers-accelerate-computer-vision-accuracy-and-improve-3d-scanning-models.aspx)
介绍:，第28届IEEE计算机视觉与模式识别(CVPR)大会在美国波士顿举行。微软研究员们在大会上展示了比以往更快更准的计算机视觉图像分类新模型，并介绍了如何使用Kinect等传感器实现在动态或低光环境的快速大规模3D扫描技术.
* [《Machine Learning for Humans》](https://github.com/marcotcr/mlforhumans)
介绍:(文本)机器学习可视化分析工具.
* [《A Plethora of Tools for Machine Learning》](http://knowm.org/machine-learning-tools-an-overview/)
介绍:机器学习工具包/库的综述/比较.
* [《The art of visualizing visualizations: a best practice guide》](http://sapblog.be/en/the-art-of-visualizing-visualizations-a-best-practice-guide/)
介绍:数据可视化最佳实践指南.
* [《MIT Machine Learning for Big Data and Text Processing Class Notes – Day 1》](http://blog.adnanmasood.com/2015/06/08/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-1/)
介绍:[Day 1](http://blog.adnanmasood.com/2015/06/08/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-1/)、[Day 2](http://blog.adnanmasood.com/2015/06/09/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-2/)、[Day 3](http://blog.adnanmasood.com/2015/06/11/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-3/)、[Day 4](http://blog.adnanmasood.com/2015/06/12/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-4/)、[Day 5](http://blog.adnanmasood.com/2015/06/12/mit-machine-learning-for-big-data-and-text-processing-class-notes-day-5/).
* [《Getting “deep” about “deep learning”》](http://whatsnext.nuance.com/in-the-labs/what-is-deep-machine-learning/)
介绍:深度学习之“深”——DNN的隐喻分析.
* [《Mixture Density Networks》](http://blog.otoro.net/2015/06/14/mixture-density-networks/)
介绍:混合密度网络.
* [《Interview Questions for Data Scientist Positions》](https://medium.com/@D33B/interview-questions-for-data-scientist-positions-5ad3c5d5b8bd)
介绍:数据科学家职位面试题.
* [《Accurately Measuring Model Prediction Error》](http://scott.fortmann-roe.com/docs/MeasuringError.html)
介绍:准确评估模型预测误差.
* [《Continually updated Data Science Python Notebooks》](https://github.com/donnemartin/data-science-ipython-notebooks)
介绍:Continually updated Data Science Python Notebooks.
* [《How to share data with a statistician》](https://github.com/jtleek/datasharing)
介绍:How to share data with a statistician.
* [《The Eyescream Project NeuralNets dreaming natural images》](http://soumith.ch/eyescream/)
介绍:来自Facebook的图像自动生成.
* [《How to share data with a statistician》](https://github.com/jtleek/datasharing)
介绍:How to share data with a statistician.
* [《A Neural Conversational Model》](http://arxiv.org/abs/1506.05869)
介绍:(Google)神经(感知)会话模型.
* [《The 50 Best Masters in Data Science》](http://www.datasciencecentral.com/profiles/blogs/the-50-best-masters-in-data-science)
介绍:The 50 Best Masters in Data Science.
* [《NLP常用信息资源》](http://forum.memect.com/thread/nlp%E5%B8%B8%E7%94%A8%E4%BF%A1%E6%81%AF%E8%B5%84%E6%BA%90/)
介绍:NLP常用信息资源.
* [《Conditional Random Fields as Recurrent Neural Networks》](http://www.robots.ox.ac.uk/~szheng/papers/CRFasRNN.pdf)
介绍:语义图像分割的实况[演示](http://www.robots.ox.ac.uk/~szheng/crfasrnndemo),通过深度学习技术和概率图模型的语义图像分割.
* [《Fully Convolutional Networks for Semantic Segmentation》](http://www.cs.berkeley.edu/~jonlong/long_shelhamer_fcn.pdf)
介绍:Caffe模型/代码:面向图像语义分割的全卷积网络,[模型代码](https://github.com/BVLC/caffe/wiki/Model-Zoo#fcn).
* [《Growing Pains for Deep Learning》](http://cacm.acm.org/news/188737-growing-pains-for-deep-learning/fulltext)
介绍:深度学习——成长的烦恼.
* [《Clustering Text Data Streams – A Tree based Approach with Ternary Function and Ternary Feature Vector 》](http://www.sciencedirect.com/science/article/pii/S1877050914005274)
介绍:基于三元树方法的文本流聚类.
* [《Foundations and Advances in Data Mining》](http://cs.ucla.edu/~wwc/course/cs245a/mining%20book.pdf)
介绍:Free Ebook:数据挖掘基础及最新进展.
* [《The Deep Learning Revolution: Rethinking Machine Learning Pipelines》](http://www.infoq.com/presentations/deep-learning)
介绍:深度学习革命.
* [《The Definitive Guide to Do Data Science for Good》](http://blog.datalook.io/definitive-guide-data-science-good/)
介绍:数据科学(实践)权威指南.
* [《Microsoft Academic Graph》](http://research.microsoft.com/en-us/projects/mag/)
介绍:37G的微软学术图谱数据集.
* [《Challenges and Opportunities Of Machine Learning In Production》](https://www.youtube.com/watch?v=UEwDwTkWwdc&hd=1)
介绍:生产环境(产品级)机器学习的机遇与挑战.
* [《Neural Nets for Newbies》](https://www.youtube.com/watch?v=Cu6A96TUy_o)
介绍:神经网络入门.
* [《A Nearly-Linear Time Framework for Graph-Structured Sparsity》](http://jmlr.org/proceedings/papers/v37/hegde15.pdf)
介绍:来自麻省理工的结构化稀疏论文.
* [《Optimal and Adaptive Algorithms for Online Boosting》](http://jmlr.org/proceedings/papers/v37/beygelzimer15.pdf)
介绍:来自雅虎的机器学习小组关于在线Boosting的论文 .
* [《Top 20 Python Machine Learning Open Source Projects》](http://www.kdnuggets.com/2015/06/top-20-python-machine-learning-open-source-projects.html)
介绍:20个最热门的开源(Python)机器学习项目.
* [《The Parallel C++ Statistical Library for Bayesian Inference: QUESO》](http://arxiv.org/abs/1507.00398)
介绍:C++并行贝叶斯推理统计库QUESO,[github code](http://libqueso.com/).
* [《《Deep learning》Yann LeCun, Yoshua Bengio, Geoffrey Hinton (2015) 》](http://www.nature.com/nature/journal/v521/n7553/full/nature14539.html)
介绍:Nature:LeCun/Bengio/Hinton的最新文章《深度学习》,Jürgen Schmidhuber的最新评论文章[《Critique of Paper by “Deep Learning Conspiracy” (Nature 521 p 436)》](http://people.idsia.ch/~juergen/deep-learning-conspiracy.html).
* [《Palladium》](https://github.com/ottogroup/palladium)
介绍:基于Scikit-Learn的预测分析服务框架Palladium.
* [《Advances in Structured Prediction》](http://hunch.net/~l2s/merged.pdf)
介绍:John Langford和Hal Daume III在ICML2015上关于Learning to Search的教学讲座幻灯片.
* [《100 open source Big Data architecture papers for data professionals》](https://www.linkedin.com/pulse/100-open-source-big-data-architecture-papers-anil-madan)
介绍:读完这100篇论文 就能成大数据高手,[国内翻译](http://www.csdn.net/article/2015-07-07/2825148/1).
* [《Social Media & Text Analytics》](http://socialmedia-class.org/syllabus.html)
介绍:NLP课程《社交媒体与文本分析》精选阅读列表.
* [《Machine Learning for Developers》](http://xyclade.github.io/MachineLearning/)
介绍:写给开发者的机器学习指南.
* [《Hot news detection using Wikipedia》](http://hameddaily.blogspot.com/2015/06/hot-news-detection-using-wikipedia_29.html)
介绍:基于维基百科的热点新闻发现.
* [《Harvard Intelligent Probabilistic Systems Group》](https://github.com/HIPS)
介绍:(Harvard)HIPS将发布可扩展/自动调参贝叶斯推理神经网络.
* [《An Empirical Exploration of Recurrent Network Architectures》](http://jmlr.org/proceedings/papers/v37/jozefowicz15.html)
介绍:面向上下文感知查询建议的层次递归编解码器.
* [《Efficient Training of LDA on a GPU by Mean-for-Mode Estimation》](http://jmlr.org/proceedings/papers/v37/tristan15.html)
介绍:GPU上基于Mean-for-Mode估计的高效LDA训练.
* [《From the Lab to the Factory: Building a Production Machine Learning Infrastructure》](https://www.youtube.com/watch?v=v-91JycaKjc&hd=1)
介绍:从实验室到工厂——构建机器学习生产架构.
* [《6 Useful Databases to Dig for Data (and 100 more)》](http://piktochart.com/6-useful-databases-to-dig-for-data/)
介绍:适合做数据挖掘的6个经典数据集(及另外100个列表).
* [《Deep Networks for Computer Vision at Google – ILSVRC2014》](http://www.computervisiontalks.com/deep-networks-for-computer-vision-at-google/)
介绍:Google面向机器视觉的深度学习.
* [《How to choose a machine learning API to build predictive apps》](https://medium.com/@louisdorard/developer-considerations-for-choosing-a-machine-learning-api-20e2de15eb3a)
介绍:构建预测类应用时如何选择机器学习API.
* [《Exploring the shapes of stories using Python and sentiment APIs》](https://indico.io/blog/plotlines/)
介绍:Python+情感分析API实现故事情节(曲线)分析.
* [《Movie selection using R》](http://melodywolk.com/2015/07/21/movie-selection-using-r/)
介绍:(R)基于Twitter/情感分析的口碑电影推荐,此外推荐[分类算法的实证比较分析](http://freakonometrics.hypotheses.org/20002).
* [《A Tutorial on Graph-based Semi-Supervised Learning Algorithms for NLP》](http://graph-ssl.wdfiles.com/local–files/blog%3A_start/graph_ssl_acl12_tutorial_slides_final.pdf)
介绍:CMU(ACL 2012)(500+页)面向NLP基于图的半监督学习算法.
* [《Arbitrariness of peer review: A Bayesian analysis of the NIPS experiment》](http://arxiv.org/abs/1507.06411)
介绍:从贝叶斯分析NIPS，看同行评审的意义.
* [《Basics of Computational Reinforcement Learning》](http://videolectures.net/rldm2015_littman_computational_reinforcement/)
介绍:(RLDM 2015)计算强化学习入门.
* [《Deep Reinforcement Learning》](http://videolectures.net/rldm2015_silver_reinforcement_learning/)
介绍:David Silver的深度强化学习教程.
* [《On Explainability of Deep Neural Networks》](http://blog.adnanmasood.com/2015/07/31/on-explainability-of-deep-neural-networks/)
介绍:深度神经网络的可解释性.
* [《The Essential Spark Cheat Sheet》](http://info.mapr.com/rs/mapr/images/rd204-010d-spark_0.pdf)
介绍:Spark快速入门.
* [《Machine Learning for Sports and Real Time Predictions》](http://www.thetalkingmachines.com/blog/2015/7/30/machine-learning-for-sports-and-real-time-predictions)
介绍:TalkingMachines:面向体育/政治和实时预测的机器学习.
* [《CS224W: Social and Information Network Analysis Autumn 2014》](http://web.stanford.edu/class/cs224w/index.html)
介绍:Stanford社交网络与信息网络分析课程[资料](http://web.stanford.edu/class/cs224w/handouts.html)+[课设](http://web.stanford.edu/class/cs224w/projects.html)+[数据](http://web.stanford.edu/class/cs224w/resources.html).
* [《RL Course by David Silver》](https://www.youtube.com/playlist?list=PL5X3mDkKaJrL42i_jhE4N-p6E2Ol62Ofa)
介绍:David Silver(DeeMind)的强化学习课程,[slide](http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching.html).
* [《Faster deep learning with GPUs and Theano》](http://blog.dominodatalab.com/gpu-computing-and-deep-learning/)
介绍:基于Theano/GPU的高效深度学习.
* [《Introduction to R Programming》](https://www.edx.org/course/introduction-r-programming-microsoft-dat204x)
介绍:来自微软的<R编程入门>.
* [《Golang:Web Server For Performing Sentiment Analysis》](https://github.com/cdipaolo/sentiment-server)
介绍:(Go)情感分析API服务Sentiment Server.
* [《A Beginner’s Guide to Restricted Boltzmann Machines》](http://deeplearning4j.org/restrictedboltzmannmachine.html)
介绍:受限波尔兹曼机初学者指南.
* [《KDD2015十年最佳论文》](http://www.kdd.org/kdd2015/program.html)
介绍:[Mining and Summarizing Customer Reviews ](http://www.cs.uic.edu/~liub/publications/kdd04-revSummary.pdf),[Mining High-Speed Data Streams](http://homes.cs.washington.edu/~pedrod/papers/kdd00.pdf),[Optimizing Search Engines using Clickthrough Data](http://www.cs.cornell.edu/people/tj/publications/joachims_02c.pdf).
* [《Nvidia Deep Learning Courses》](http://www.hellophp.cn/archives/733)
介绍:Nvidia深度学习课程.
* [《Neural Networks and Deep Learning》](http://neuralnetworksanddeeplearning.com/)
介绍:神经网络与深度学习课程.
* [《Deep Learning Summer School 2015》](https://sites.google.com/site/deeplearningsummerschool/)
介绍:2015年深度学习暑期课程，推荐[讲师主页](http://www.iro.umontreal.ca/~memisevr).
* [《百度深度学习的图像识别进展》](http://www.cvrobot.net/image-recognition-progression-based-on-deep-learning-by-baidu/)
介绍:这是一篇关于百度文章[《基于深度学习的图像识别进展：百度的若干实践》](http://www.ccf.org.cn/sites/ccf/xhdtnry.jsp?contentId=2857471255804)的摘要,建议两篇文章结合起来阅读.
* [《Machine Learning Methods in Video Annotation》](http://rnd.azoft.com/machine-learning-methods-video-annotation/)
介绍:视频标注中的机器学习技术.
* [《Training Recurrent Neural Networks》](http://www.cs.utoronto.ca/~ilya/pubs/ilya_sutskever_phd_thesis.pdf)
介绍:博士论文:(Ilya Sutskever)RNN训练.
* [《On Explainability of Deep Neural Networks》](http://blog.adnanmasood.com/2015/07/31/on-explainability-of-deep-neural-networks/)
介绍:深度神经网络的灰色区域：可解释性问题，[中文版](http://www.csdn.net/article/2015-08-17/2825471).
* [《Machine Learning Libraries in GoLang by Category》](http://www.fodop.com/ar-1002)
介绍:Golang 实现的机器学习库资源汇总.
* [《A Statistical View of Deep Learning》](http://blog.shakirm.com/wp-content/uploads/2015/07/SVDL.pdf)
介绍:深度学习的统计分析.
* [《Deep Learning For NLP – Tips And Techniques》](http://www.researchgate.net/publication/279853751_DEEP_LEARNING_FOR_NLP_-_TIPS_AND_TECHNIQUES)
介绍:面向NLP的深度学习技术与技巧.
* [《CrowdFlower Competition Scripts: Approaching NLP》](http://blog.kaggle.com/2015/08/18/crowdflower-scripts-approaching-nlp/)
介绍:Kaggle’s CrowdFlower竞赛NLP代码集锦.
* [《CS224U: Natural Language Understanding》](http://web.stanford.edu/class/cs224u/index.html)
介绍:斯坦福的自然语言理解课程.
* [《Deep Learning and Shallow Learning》](http://freemind.pluskid.org/machine-learning/deep-learning-and-shallow-learning/)
介绍:Deep Learning与Shallow Learning 介绍
* [《A First Encounter with Machine Learning》](http://www.ics.uci.edu/~welling/teaching/ICS273Afall11/IntroMLBook.pdf)
介绍:这是一本机器学习的电子书,作者[Max Welling](http://www.ics.uci.edu/~welling/)先生在机器学习教学上面有着丰富的经验,这本书小但精致.
* [《Click Models for Web Search》](http://clickmodels.weebly.com/uploads/5/2/2/5/52257029/mc2015-clickmodels.pdf)
介绍:由荷兰阿姆斯特丹大学 & 谷歌瑞士著.
* [《Hinton CSC321课程/Deep Learning/Notes on CNN/Python/Theano/CUDA/OpenCV/…》](http://www.cnblogs.com/shouhuxianjian/p/4529235.html)
介绍:介绍个乐于总结和翻译机器学习和计算机视觉类资料的博客,包含的内容：Hinton的CSC321课程的总结；Deep Learning综述；Notes on CNN的总结；python的原理总结；Theano基础知识和练习总结；CUDA原理和编程；OpenCV一些总结.
* [《Which Algorithm Family Can Answer My Question?》](http://blogs.technet.com/b/machinelearning/archive/2015/09/01/which-algorithm-family-can-answer-my-question.aspx)
介绍:针对具体问题(应用场景)如何选择机器学习算法(系列).
* [《Free Data Science Books》](http://www.learndatasci.stfi.re/free-books/)
介绍:数据科学免费书分类集合
* [《Tutorial 4: Deep Learning for Speech Generation and Synthesis》](http://www.superlectures.com/iscslp2014/tutorial-4-deep-learning-for-speech-generation-and-synthesis)
介绍:深度学习在语音合成最新进展有哪些？推荐MSRA的Frank Soong老师关于语音合成的深度学习方法的录像和幻灯片与以及谷歌的LSTM-RNN合成[介绍](http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/42624.pdf),[论文](http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43893.pdf)
* [《The Art of Data Science》](https://leanpub.com/artofdatascience)
介绍:新书(可免费下载):数据科学的艺术
* [《Pattern Recognition and Machine Learning》](http://research.microsoft.com/en-us/um/people/cmbishop/prml/)
介绍:模式识别与机器学习书籍推荐,[中文版](https://www.dropbox.com/s/sx95jq7n7zerrjl/PRML_Translation.pdf?dl=0) or [备份](http://pan.baidu.com/s/1hqheD5E)
* [《an introduction to visualizing DATA》](http://piksels.com/wp-content/uploads/2009/01/visualizingdata.pdf)
介绍:数据可视化介绍(23页袖珍小册子)
* [《That’s So Annoying!!!: A Lexical and Frame-Semantic Embedding Based Data Augmentation Approach to Automatic Categorization of Annoying Behaviors using #petpeeve Tweets ∗》](https://www.cs.cmu.edu/~yww/papers/emnlp2015petpeeves.pdf)
介绍:这篇论文荣获EMNLP2015的最佳数据/资源奖优秀奖,[标注的推特数据集](https://www.cs.cmu.edu/~yww/data/petpeeves.zip)
* [《26 Things I Learned in the Deep Learning Summer School》](http://www.marekrei.com/blog/26-things-i-learned-in-the-deep-learning-summer-school/)
介绍:作者在深度学习的思考.
* [《Data-Visualization Tools & Books》](http://keshif.me/demo/VisTools)
介绍:数据可视化常用工具软件资源汇总
* [《Machine Learning and Probabilistic Graphical Models Course》](http://www.cedar.buffalo.edu/~srihari/CSE574/)
介绍:Buffalo大学教授Sargur Srihari的“机器学习和概率图模型”的视频课程
* [《Understanding Machine Learning: From Theory to Algorithms》](http://www.cs.huji.ac.il/~shais/UnderstandingMachineLearning/index.html)
介绍:耶路撒冷希伯来大学教授Shai Shalev-Shwartz和滑铁卢大学教授Shai Ben-David的新书Understanding Machine Learning: From Theory to Algorithms,此书写的比较偏理论，适合对机器学习理论有兴趣的同学选读
* [《Machine Learning Checklist》](http://machinelearningmastery.com/machine-learning-checklist/)
介绍:机器学习学习清单
* [《Neural Networks and Deep Learning》](http://neuralnetworksanddeeplearning.com/)
介绍:Michael Nielsen的免费在线电子书”Neural Networks and Deep Learning”,深入浅出，但内容涵盖更广：从经典的NN到CNN，从BP到梯度消失问题都有所覆盖，还有示范代码
* [《NLP界有哪些神级人物？》](http://www.zhihu.com/question/32318281)
介绍:知乎上面的一篇关于NLP界有哪些神级人物？提问。首推Michael Collins
* [《机器学习温和指南》](http://www.csdn.net/article/2015-09-08/2825647)
介绍:机器学习与NLP专家、MonkeyLearn联合创始人&CEO Raúl Garreta面向初学者大体概括使用机器学习过程中的重要概念，应用程序和挑战，旨在让读者能够继续探寻机器学习知识。
* [《Gradient Boosted Regression Trees》](http://nbviewer.ipython.org/github/pprett/pydata-gbrt-tutorial/blob/master/gbrt-tutorial.ipynb)
介绍:(IPN)基于Scikit-Learn的GBRT(Gradient Boost Regression Tree)教程，[slide](http://orbi.ulg.ac.be/bitstream/2268/163521/1/slides.pdf)
* [《Apache SINGA : Distributed Deep Learning System》](http://www.comp.nus.edu.sg/~dbsystem/singa/)
介绍: 无需做深度学习就能用的分布式深度学习软件.
* [《E-commerce Recommendation with Personalized Promotion》](http://dl.acm.org/citation.cfm?id=2800178)
介绍: 在亚马逊数据和众包Mechanical Turk上，实现了来自彩票和拍卖的机制，以收集用户对产品的乐意购买价格（WTP，willingness-to-pay）训练集。 E-commerce Recommendation with Personalized Promotion [Zhao,RecSys15] 回归模型预测未知WTP，提升卖家利润和消费者满意度
* [《Scalable Machine Learning》](https://www.edx.org/course/scalable-machine-learning-uc-berkeleyx-cs190-1x)
介绍:来自伯克利分校的大规模机器学习.
* [《机器学习资料大汇总》](http://www.52ml.net/star)
介绍:来自52ml的机器学习资料大汇总.
* [《Automatic Summarization》](http://www.cis.upenn.edu/~nenkova/1500000015-Nenkova.pdf)
介绍:这本书的作[者McKeown](http://www.cis.upenn.edu/~nenkova/)是2013年世界首个数据科学院（位于哥伦比亚大学）主任，她亦是ACL、AAAI和ACM Fellow .
* [《Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing》](http://www.emnlp2015.org/proceedings/EMNLP/index.html)
介绍:EMNLP-15文本摘要若干.
* [《Recommender Systems (Machine Learning Summer School 2014 @ CMU)》](http://www.slideshare.net/xamat/recommender-systems-machine-learning-summer-school-2014-cmu)
介绍:来自Netflix的Xavier Amatriain在Summer School 2014 @ CMU上长达4小时的报告，共248页，是对推荐系统发展的一次全面综述，其中还包括Netflix在个性化推荐方面的一些经验介绍.
* [《BigData Stream Mining》](http://www.ecmlpkdd2015.org/sites/default/files/ECMLPKDD2015Slides.pdf)
介绍:(ECML PKDD 2015)大数据流挖掘教程,此外推荐[ECML PKDD 2015 Tutorial列表](http://www.ecmlpkdd2015.org/program/tutorial-list).
* [《Deep learning on Spark with Keras》](https://github.com/maxpumperla/elephas)
介绍:Spark上的Keras深度学习框架Elephas.
* [《Prof. Surya Ganguli – The statistical physics of deep learning》](https://www.youtube.com/watch?v=7KCWcx-YIRI&hd=1)
介绍:Surya Ganguli深度学习统计物理学.
* [《(系统/算法/机器学习/深度学习/图模型/优化/…)在线视频课程列表》](http://cmlakhan.github.io/courses/videos.html)
介绍:(系统/算法/机器学习/深度学习/图模型/优化/…)在线视频课程列表.
* [《Introduction to Topic Modeling in Python》](http://chdoig.github.io/pytexas2015-topic-modeling/)
介绍:(PyTexas 2015)Python主题建模.
* [《Large Scale Distributed Deep Learning on Hadoop Clusters》](http://yahoohadoop.tumblr.com/post/129872361846/large-scale-distributed-deep-learning-on-hadoop/)
介绍:Hadoop集群上的大规模分布式机器学习.
* [《Top Deep Learning Employers Based On LinkedIn Data》](http://www.vordot.com/deep-learning-employers-w-12020/)
介绍:基于LinkedIn数据得出的深度学习热门”东家”排行.
* [《Neural Net in C++ Tutorial》](https://vimeo.com/19569529)
介绍:(c++)神经网络手把手实现教程.
* [《Large-scale CelebFaces Attributes (CelebA) Dataset》](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html)
介绍:香港中文大学汤晓鸥教授实验室公布的大型人脸识别数据集： Large-scale CelebFaces Attributes (CelebA) Dataset 10K 名人，202K 脸部图像，每个图像40余标注属性.
* [《Unsupervised Feature Learning in Computer Vision》](https://www.cs.nyu.edu/web/Research/Theses/goroshin_ross.pdf)
介绍:面向机器视觉的无监督特征学习,[Ross Goroshin’s webpage](https://cs.nyu.edu/~goroshin/).
* [《Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks》](http://arxiv.org/pdf/1506.03099v3.pdf)
介绍:谷歌研究院Samy Bengio等人最近写的RNN的Scheduled Sampling训练方法论文.
* [《Essential Machine Learning Algorithms in a nutshell》](https://manish.wordpress.com/2015/10/02/essential-machine-learning-algorithms-in-a-nutshell/)
介绍:机器学习基本算法简要入门.
* [《A Huge List of Machine Learning And Statistics Repositories》](http://blog.josephmisiti.com/a-huge-list-of-machine-learning-repositories/)
介绍:Github机器学习/数学/统计/可视化/深度学习相关项目大列表.
* [《Information Processing and Learning》](http://www.cs.cmu.edu/~aarti/Class/10704_Spring15/lecs.html)
介绍:CMU的信息论课程.
* [《Scheduled sampling for sequence prediction with recurrent neural networks》](http://arxiv.org/pdf/1506.03099v3.pdf)
介绍:谷歌研究院[Samy Bengio](http://bengio.abracadoudou.com/)等人最近写的RNN的Scheduled Sampling训练方法论文.
* [《基于Hadoop集群的大规模分布式深度学习》](http://www.csdn.net/article/2015-10-01/2825840)
介绍:基于Hadoop集群的大规模分布式深度学习.
* [《Learning both Weights and Connections for Efficient Neural Networks习》](http://arxiv.org/abs/1506.02626)
介绍:来自斯坦福大学及NVIDIA的工作，很实在很实用。采用裁剪网络连接及重训练方法，可大幅度减少CNN模型参数。针对AlexNet、VGG等模型及ImageNet数据，不损失识别精度情况下，模型参数可大幅度减少9-13倍.
* [《Apache Singa –A General Distributed Deep Learning Platform》](http://www.comp.nus.edu.sg/~dbsystem/singa/)
介绍:无需做深度学习就能用的分布式深度学习软件,[github](https://github.com/apache/incubator-singa).
* [《24 Ultimate Data Scientists To Follow in the World Today》](http://www.analyticsvidhya.com/blog/2015/09/ultimate-data-scientists-world-today/)
介绍:当今世界最NB的25位大数据科学家,通过他们的名字然后放在google中搜索肯定能找到很多很棒的资源[译文](http://blog.csdn.net/heyongluoyao8/article/details/48598169).
* [《Deep Learning for NLP – Lecture October 2015》](https://github.com/nreimers/deeplearning4nlp-tutorial/tree/master/2015-10_Lecture/)
介绍:Nils Reimers面向NLP的深度学习(Theano/Lasagne)系列教程.
* [《Connection between probability theory and real analysis》](https://ccle.ucla.edu/mod/page/view.php?id=834267)
介绍:主讲人是[陶哲轩](https://ccle.ucla.edu/mod/page/view.php?id=834267),资料[Probability: Theory and Examples](http://www.math.duke.edu/~rtd/PTE/PTE4_1.pdf),[笔记](https://terrytao.wordpress.com/category/275a-probability-theory/).
* [《Data Science Learning Resources》](http://www.districtdatalabs.com/#!resources/c21hq)
介绍:数据科学(学习)资源列表.
* [《8 Tactics to Combat Imbalanced Classes in Your Machine Learning Dataset》](http://machinelearningmastery.com/tactics-to-combat-imbalanced-classes-in-your-machine-learning-dataset/)
介绍:应对非均衡数据集分类问题的八大策略.
* [《Top 20 Data Science MOOCs》](https://datarithms.wordpress.com/2015/08/16/top-20-data-science-moocs/)
介绍:重点推荐的20个数据科学相关课程.
* [《Recurrent Neural Networks》](https://shapeofdata.wordpress.com/2015/10/20/recurrent-neural-networks/)
介绍:递归神经网络.
* [《Histograms of Oriented Gradients》](http://www.cs.duke.edu/courses/fall15/compsci527/notes/hog.pdf)
介绍:(HOG)学习笔记.
* [《Computational modelling courses》](http://aidanhorner.blogspot.co.uk/2015/10/computational-modelling-courses.html)
介绍:计算建模/计算神经学课程汇总.
* [《How We Use Deep Learning to Classify Business Photos at Yelp》](http://engineeringblog.yelp.com/2015/10/how-we-use-deep-learning-to-classify-business-photos-at-yelp.html)
介绍:(Yelp)基于深度学习的商业图片分类.
* [《Neural Networks and Deep Learning》](http://neuralnetworksanddeeplearning.com/)
介绍:免费在线书《Neural Networks and Deep Learning》神经网络与深度学习。目前提供了前四章的草稿，[第一章](http://mp.weixin.qq.com/s?__biz=MzIxMjAzNDY5Mg==&mid=400067748&idx=1&sn=9c88eadfba5462281cd496e85ba3329c)通过手写数字识别的例子介绍NN，第二章讲反向传播算法，第三章讲反向传播算法的优化，第四章讲NN为什么能拟合任意函数。大量python代码例子和交互动画，生动有趣.
* [《Books to Read if You Might Be Interested in Data Science》](http://www.datasciguide.com/books-to-read-if-you-might-be-interested-in-data-science/)
介绍:数据科学大咖荐书(入门).
* [《Deep Learning for NLP resources》](https://github.com/andrewt3000/DL4NLP)
介绍:NLP 深度学习资源列表.
* [《GitXiv》](http://gitxiv.com/)
介绍:很多arXiv上面知名论文可以在这个网站找到github的项目链接.
* [《Learning Multi-Domain Convolutional Neural Networks for Visual Tracking》](http://arxiv.org/pdf/1510.07945v1.pdf)
介绍:深度学习在视觉跟踪的探索.
* [《Beginners Guide: Apache Spark Machine Learning Scenario With A Large Input Dataset》](http://fullstackml.com/2015/10/29/beginners-guide-apache-spark-machine-learning-scenario-with-a-large-input-dataset/)
介绍:Spark机器学习入门实例——大数据集(30+g)二分类.
* [《Semantic Scholar》](https://www.semanticscholar.org/)
介绍:保罗艾伦人工智能实验室表示，Google Scholar是十年前的产物，他们现在想要做进一步的提高。于是推出了全新的，专门针对科学家设计的学术搜索引擎Semantic Scholar.
* [《Semi-Supervised Learning》](http://www.acad.bg/ebook/ml/MITPress-%20SemiSupervised%20Learning.pdf)
介绍:半监督学习,Chapelle.篇篇都是经典,作者包括Vapnik,Bengio,Lafferty,Jordan.此外推荐[Xiaojin (Jerry) Zhu](http://pages.cs.wisc.edu/~jerryzhu/)编写的[Introduction to Semi-Supervised Learning](http://www.morganclaypool.com/doi/abs/10.2200/S00196ED1V01Y200906AIM006).
介绍:Spark机器学习入门实例——大数据集(30+g)二分类.
* [《Free Resources for Beginners on Deep Learning and Neural Network》](http://www.analyticsvidhya.com/blog/2015/11/free-resources-beginners-deep-learning-neural-network/)
介绍:为入门者准备的深度学习与神经网络免费资源.
* [《TensorFlow is an Open Source Software Library for Machine Intelligence》](http://tensorflow.org/)
介绍:Google 开源最新机器学习系统 TensorFlow,此外提供TensorFlow白皮书[white paper of tensorflow 2015](http://pan.baidu.com/s/1jGyFPki).[hacker news](https://news.ycombinator.com/item?id=10532957),[Google大牛解读TensorFlow](https://www.youtube.com/watch?v=90-S1M7Ny_o&t=21m2s)
* [《Veles:Distributed machine learning platform》](https://github.com/samsung/veles)
介绍:三星开源的快速深度学习应用程序开发分布式平台.
* [《DMTK:Microsoft Distributed Machine Learning Tookit 》](https://github.com/Microsoft/DMTK)
介绍:分布式机器学习工具包.
* [《Semantics Approach to Big Data and Event Processing》](http://wiki.knoesis.org/index.php/BigDataTutorial)
介绍:语义大数据——大数据/事件处理的语义方法.
* [《LSTM(Long Short Term Memory)和RNN(Recurrent)学习教程》](http://www.zhihu.com/question/29411132)
介绍:LSTM(Long Short Term Memory)和RNN(Recurrent)学习教程.
* [《Marvin：A minimalist GPU-only N-dimensional ConvNet framework》](http://marvin.is/)
介绍:Princeton Vision Group的深度学习库开源.
* [《Ufora is a compiled, automatically parallel subset of python for data science and numerical computing》](http://ufora.github.io/ufora/)
介绍:基于AWS的自动分布式科学计算库Ufora,[Why I Open Sourced Five Years of Work](https://medium.com/art-marketing/why-i-open-sourced-five-years-of-work-c5b5e0e38a6d).
* [《Deep Learning and Deep Data Science – PyCon SE 2015》](https://www.youtube.com/watch?v=wBKfGaakFp8&hd=1)
介绍:(PyCon SE 2015)深度学习与深度数据科学.
* [《Zhi-Hua Zhou Papers》](https://scholar.google.com/citations?user=rSVIHasAAAAJ&hl=zh-CN&oi=ao)
介绍:推荐南京大学机器学习与数据挖掘研究所所长——周志华教授的Google学术主页.
* [《Advanced Linear Models for Data Science》](https://leanpub.com/lm)
介绍:免费书:面向数据科学的高级线性模型.
* [《Net2Net: Accelerating Learning via Knowledge Transfer》](http://arxiv.org/abs/1511.05641)
介绍:基于知识迁移的神经网络高效训练Net2Net.
* [《徐亦达机器学习课程 Variational Inference》](https://www.youtube.com/playlist?list=PLFze15KrfxbF0n1zTNoFIaDpxnSyfgNgc)
介绍:徐亦达机器学习课程 Variational Inference.
* [《Learning the Architecture of Deep Neural Networks》](http://arxiv.org/abs/1511.05497v1)
介绍:深度神经网络结构学习.
* [《Multimodal Deep Learning》](http://ai.stanford.edu/~ang/papers/icml11-MultimodalDeepLearning.pdf)
介绍:来自斯坦福大学的Multimodal Deep Learning papers.
* [《深度学习简析,TensorFlow,Torch,Theano,Mxnet》](http://chiffon.gitcafe.io/2015/11/16/long.html)
介绍:深度学习简析,TensorFlow,Torch,Theano,Mxnet.
* [《”Notes Essays —CS183C: Technology-enabled Blitzscaling — Stanford University》](https://medium.com/notes-essays-cs183c-technology-enabled-blitzscalin/latest)
介绍:这个专栏是一个stanford学生做的CS183c课程的一个note，该课程是由Reid Hoffman等互联网boss级人物开设的，每节课请一位巨头公司的相关负责人来做访谈，讲述该公司是怎么scale的。最新两期分别请到了雅虎的梅姐和airbnb创始人Brian Chesky。.
* [《Natural Language Understanding with Distributed Representation》](https://github.com/nyu-dl/NLP_DL_Lecture_Note)
介绍:基于分布式表示的自然语言理解(100+页),[论文](http://arxiv.org/abs/1511.07916).
* [《Recommender Systems Handbook》](http://link.springer.com/book/10.1007/978-1-4899-7637-6)
介绍:推荐系统手册.
* [《Understanding LSTM Networks》](http://colah.github.io/posts/2015-08-Understanding-LSTMs/index.html)
介绍:理解LSTM网络[翻译](http://www.csdn.net/article/2015-11-25/2826323).
* [《Machine Learning at Quora》](https://www.linkedin.com/pulse/machine-learning-quora-xavier-amatriain)
介绍:机器学习在quora中的应用.
* [《On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models》](http://arxiv.org/abs/1511.09249)
介绍:思维学习——RL+RNN算法信息论.
* [《The 5 Ways Data Scientists Keep Learning After College》](https://blog.rjmetrics.com/2015/12/01/the-5-ways-data-scientists-keep-learning-after-college/)
介绍:数据科学家毕业后继续学习的5种方式.
* [《Deep Learning in Neural Networks: An Overview》](http://arxiv.org/abs/1404.7828)
介绍:深度学习在神经网络的应用.
* [《Contextual Learning》](http://arxiv.org/abs/1511.06429)
介绍:上下文学习,[代码](https://gitlab.tubit.tu-berlin.de/rbo-lab/concarne).
* [《Machine Learning For Complete Beginners》](http://pythonforengineers.com/machine-learning-for-complete-beginners/)
介绍:机器学习零基础入门,[代码](https://github.com/shantnu/Titanic-Machine-Learning).
* [《2015年中国计算机学会（CCF）优秀博士学位论文》](http://www.ccf.org.cn/sites/ccf/xhdtnry.jsp?contentId=2897719129810)
介绍:2015年度CCF优秀博士学位论文奖论文列表.
* [《Learning to Hash Paper, Code and Dataset》](http://cs.nju.edu.cn/lwj/L2H.html)
介绍:Learning to Hash Paper, Code and Dataset.
* [《Neural networks with Theano and Lasagne》](https://www.youtube.com/watch?v=dtGhSE1PFh0)
介绍:(PyData2015)基于Theano/Lasagne的CNN/RNN教程,[github](https://github.com/ebenolson/pydata2015).
* [《神经网络与深度学习讲义》](http://vdisk.weibo.com/s/ayG13we2ltDAT)
介绍:复旦大学[邱锡鹏](http://weibo.com/xpqiu)老师编写的神经网络与深度学习讲义,[ppt](http://vdisk.weibo.com/s/ayG13we2lDzcV).
* [《Microsoft Open Sources Distributed Machine Learning Toolkit》](http://www.dmtk.io/index.html)
介绍:微软亚洲研究院开源分布式机器学习工具包.
* [《语音识别的技术原理是什么？》](https://www.zhihu.com/question/20398418)
介绍:语音识别的技术原理浅析
* [《Michael I. Jordan》](http://www.cs.berkeley.edu/~jordan/)
介绍:迈克尔·I.乔丹的主页.根据主页可以找到很多资源。迈克尔·I.乔丹是知名的计算机科学和统计学学者，主要研究机器学习和人工智能。他的重要贡献包括指出了机器学习与统计学之间的联系，并推动机器学习界广泛认识到贝叶斯网络的重要性。
* [《Geoff Hinton》](http://www.cs.toronto.edu/~hinton/)
介绍:杰弗里·埃弗里斯特·辛顿 FRS是一位英国出生的计算机学家和心理学家，以其在神经网络方面的贡献闻名。辛顿是反向传播算法和对比散度算法的发明人之一，也是深度学习的积极推动者.通过他的主页可以发掘到很多Paper以及优秀学生的paper,此外推荐他的学生[Yann Lecun](http://yann.lecun.com/)主页
* [《Yoshua Bengio》](http://www.iro.umontreal.ca/~bengioy/yoshua_en/index.html)
介绍:Yoshua Bengio是机器学习方向的牛人,如果你不知道可以阅读[对话机器学习大神Yoshua Bengio（上）](http://www.infoq.com/cn/articles/ask-yoshua-bengio),[对话机器学习大神Yoshua Bengio（下）](http://www.infoq.com/cn/articles/ask-yoshua-bengio-2)
* [《Large Scale Deep Learning within google》](http://static.googleusercontent.com/media/research.google.com/en/us/people/jeff/CIKM-keynote-Nov2014.pdf)
介绍:google大规模深度学习应用演进
* [《Deep Learning: An MIT Press Book in Preparation》](http://goodfeli.github.io/dlbook/)
介绍:MIT出版的深度学习电子书,公开电子书
* [《A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction》](http://arxiv.org/abs/1512.06293)
介绍:深度卷积神经网络(CNN)提取特征的数学理论
* [《Microsoft Research Asia：Kaiming He》](http://research.microsoft.com/en-us/um/people/kahe/)
介绍:推荐微软亚洲研究院何恺明主页
* [《Speech and Language Processing (3rd ed. draft)》](http://web.stanford.edu/~jurafsky/slp3/)
介绍:《语音与语言处理》第三版(草稿)
* [《LSA 311: Computational Lexical Semantics – Summer 2015》](http://web.stanford.edu/~jurafsky/li15/)
介绍:Stanford新课”计算词汇语义学”
* [《上海交大张志华老师的统计机器学习与机器学习导论视频》](http://ocw.sjtu.edu.cn/G2S/OCW/cn/CourseDetails.htm?Id=397)
介绍:上海交大张志华老师的统计机器学习与机器学习导论视频[链接:](http://pan.baidu.com/s/1mgPi7jU )密码: r9ak .[概率基础](http://ocw.sjtu.edu.cn/G2S/OCW/cn/CourseDetails.htm?Id=398)
* [《Computational Linguistics and Deep Learning》](http://www.mitpressjournals.org/doi/pdf/10.1162/COLI_a_00239)
介绍:computational linguistics and deep learning[视频](http://techtalks.tv/talks/computational-linguistics-and-deep-learning/61759/),推荐[Deep Learning: An Introduction from the NLP Perspective](https://speakerdeck.com/baojie/deep-learning-an-introduction-from-the-nlp-perspective-by-kevin-duh)
* [《Black Hat USA 2015 – Deep Learning On Disassembly》](https://www.youtube.com/watch?v=zfVfpMcUkq8)
介绍:(BlackHat2015)深度学习应用之流量鉴别(协议鉴别/异常检测),[slide])(https://www.blackhat.com/docs/us-15/materials/us-15-Wang-The-Applications-Of-Deep-Learning-On-Traffic-Identification.pdf),[material](https://www.blackhat.com/docs/us-15/materials/us-15-Wang-The-Applications-Of-Deep-Learning-On-Traffic-Identification-wp.pdf)
* [《LibRec：A Java Library for Recommender Systems》](http://www.librec.net/)
介绍:一个推荐系统的Java库
* [《Multi-centrality Graph Spectral Decompositions and their Application to Cyber Intrusion Detection》](http://arxiv.org/abs/1512.07372)
介绍:多中心图的谱分解及其在网络入侵检测中的应用(MC-GPCA&MC-GDL)
* [《Computational Statistics in Python》](http://people.duke.edu/~ccc14/sta-663/)
介绍:用Python学计算统计学
* [《New open-source Machine Learning Framework written in Java》](http://blog.datumbox.com/new-open-source-machine-learning-framework-written-in-java/)
介绍:datumbox-framework——Java的开源机器学习框架，该框架重点是提供大量的机器学习算法和统计检验，并能够处理中小规模的数据集
* [《Awesome Recurrent Neural Networks》](http://jiwonkim.org/awesome-rnn/)
介绍:递归神经网络awesome系列,涵盖了书籍,项目,paper等
* [《Pedro Domingos》](http://homes.cs.washington.edu/~pedrod/)
介绍:Pedro Domingos是华盛顿大学的教授,主要研究方向是机器学习与数据挖掘.在2015年的ACM webinar会议,曾发表了关于[盘点机器学习领域的五大流派](http://www.almosthuman.cn/2015/11/28/t8ysa/)主题演讲.他的个人主页拥有很多相关研究的paper以及他的教授课程.
* [《Video resources for machine learning》](http://dustintran.com/blog/video-resources-for-machine-learning/)
介绍:机器学习视频集锦
* [《Deep Machine Learning libraries and frameworks》](https://medium.com/@abduljaleel/deep-machine-learning-libraries-and-frameworks-5fdf2bb6bfbe#.lwn2iyjsn)
介绍:深度机器学习库与框架
* [《大数据/数据挖掘/推荐系统/机器学习相关资源》](https://github.com/Flowerowl/Big-Data-Resources)
介绍:这篇文章内的推荐系统资源很丰富,作者很有心,摘录了《推荐系统实战》内引用的论文.
* [《Bayesian Methods in Astronomy: Hands-on Statistics》](http://nbviewer.ipython.org/github/jakevdp/AAS227Workshop/blob/master/Index.ipynb)
介绍:(天文学)贝叶斯方法/MCMC教程——统计实战
* [《Statistical Learning with Sparsity: The Lasso and Generalizations》](http://web.stanford.edu/~hastie/StatLearnSparsity/index.html)
介绍:免费书:统计稀疏学习,作者[Trevor Hastie](http://web.stanford.edu/~hastie/)与[Rob Tibshirani](http://statweb.stanford.edu/~tibs/)都是斯坦福大学的教授,Trevor Hastie更是在统计学学习上建树很多
* [《The Evolution of Distributed Programming in R》](http://www.mango-solutions.com/wp/2016/01/the-evolution-of-distributed-programming-in-r/)
介绍:R分布式计算的进化,此外推荐[(R)气候变化可视化](https://aschinchon.wordpress.com/2016/01/07/climatic-change-at-a-glance/),[(R)马尔可夫链入门](http://blog.revolutionanalytics.com/2016/01/getting-started-with-markov-chains.html)
* [《neon workshop at Startup.ML: Sentiment Analysis and Deep Reinforcement Learning》](http://www.nervanasys.com/neon-workshop-at-startup-ml-sentiment-analysis-and-deep-reinforcement-learning/)
介绍:Nervana Systems在[Startup.ML](http://startup.ml/)的主题研讨会——情感分析与深度强化学习
* [《Understanding Convolution in Deep Learning》](http://timdettmers.com/2015/03/26/convolution-deep-learning/)
介绍:深度学习卷积概念详解.
* [《Python libraries for building recommender systems》](http://faroba.com/2015/12/03/a-python-libraries-for-building-recommender-systems/)
介绍:Python推荐系统开发库汇总.
* [《Neural networks class – Université de Sherbrooke》](http://info.usherbrooke.ca/hlarochelle/neural_networks/content.html)
介绍:超棒的神经网络课程，深入浅出介绍深度学习，由Hugo Larochelle（Yoshua Bengio的博士生，Geoffrey Hinton之前的博士后）主讲，强烈推荐.
* [《CS231n: Convolutional Neural Networks for Visual Recognition》](http://vision.stanford.edu/teaching/cs231n/index.html)
介绍:斯坦福新课程,面向视觉识别的卷积神经网络(Fei-Fei Li & Andrej Karpathy),[slides+video](http://vision.stanford.edu/teaching/cs231n/syllabus.html),[homework](http://cs231n.github.io/).
* [《NIPS 2015 Deep Learning Symposium Part I》](http://yanran.li/peppypapers/2015/12/11/nips-2015-deep-learning-symposium-part-i.html)
介绍:NIPS 2015会议总结第一部分,[第二部分](http://yanran.li/peppypapers/2016/01/09/nips-2015-deep-learning-symposium-part-ii.html).
* [《python机器学习入门资料梳理》](http://michaelxiang.me/2015/12/16/python-machine-learning-list/)
介绍:python机器学习入门资料梳理.
* [《Reading Text in the Wild with Convolutional Neural Networks》](http://www.robots.ox.ac.uk/~vgg/publications/2016/Jaderberg16/)
介绍:牛津大学著名视觉几何组VGG在IJCV16年首卷首期: Reading Text in the Wild with Convolutional Neural Networks,Jaderberg。这篇期刊文章融合了之前两篇会议(ECCV14,NIPS14ws)，定位和识别图片中的文本(叫text spotting)。 端到端系统: 检测Region + 识别CNN。论文、数据和代码.
* [《Yet Another Computer Vision Index To Datasets (YACVID)》](http://riemenschneider.hayko.at/vision/dataset/)
介绍:计算机视觉的一个较大的数据集索引, 包含387个标签，共收录了314个数据集合，点击标签云就可以找到自己需要的库了.
* [《Why SLAM Matters, The Future of Real-Time SLAM, and Deep Learning vs SLAM》](http://www.computervisionblog.com/2016/01/why-slam-matters-future-of-real-time.html)
介绍:Tombone 对 ICCV SLAM workshop 的总结： the future of SLAM, SLAM vs deep learning 重点介绍了 monoSLAM 和 LSD-SLAM，而且讨论了 feature-based 和 feature-free method 的长短。在全民deep learning做visual perception的时候，再来读读CV中的 geometry.
