瓜瓜、Aileen编译

超级大汇总!200多个最好的机器学习、NLP和Python教程

这篇文章包含了我目前为止找到的最好的教程内容。这不是一张罗列了所有网上跟机器学习相关教程的清单——不然就太冗长太重复了。我这里并没有包括那些质量一般的内容。我的目标是把能找到的最好的教程与机器学习自然语言处理的延伸主题们连接到一起。

我这里指的“教程”,是指那些为了简洁地传授一个概念而写的介绍性内容。我尽量避免了教科书里的章节,因为它们涵盖了更广的内容,或者是研究论文,通常对于传授概念来说并不是很有帮助。如果是那样的话,为何不直接买书呢?当你想要学习一个基本主题或者是想要获得更多观点的时候,教程往往很有用。

我把这篇文章分为了四个部分:机器学习自然语言处理,python和数学。在每个部分中我都列举了一些主题,但是因为材料的数量庞大,我不可能涉及到每一个主题。

如果你发现到我遗漏了哪些好的教程,请告诉我!我尽量把每个主题下的教程控制在五个或者六个,如果超过了这个数字就难免会有重复。每一个链接都包含了与其他链接不同的材料,或使用了不同的方式表达信息(例如:使用代码,幻灯片和长文),或者是来自不同的角度。

机器学习

Start Here with Machine Learning (machinelearningmastery.com)

https://machinelearningmastery.com/start-here/

Machine Learning is Fun! (medium.com/@ageitgey)

https://medium.com/@ageitgey/machine-learning-is-fun-80ea3ec3c471

Rules of Machine Learning: Best Practices for ML Engineering(martin.zinkevich.org)

http://martin.zinkevich.org/rules_of_ml/rules_of_ml.pdf

Machine Learning Crash Course: Part I, Part II, Part III (Machine Learning at Berkeley)

https://ml.berkeley.edu/blog/2016/11/06/tutorial-1/

https://ml.berkeley.edu/blog/2016/12/24/tutorial-2/

https://ml.berkeley.edu/blog/2017/02/04/tutorial-3/

An Introduction to Machine Learning Theory and Its Applications: A Visual Tutorial with Examples (toptal.com)

https://www.toptal.com/machine-learning/machine-learning-theory-an-introductory-primer

A Gentle Guide to Machine Learning (monkeylearn.com)

https://monkeylearn.com/blog/gentle-guide-to-machine-learning/

Which machine learning algorithm should I use? (sas.com)

https://blogs.sas.com/content/subconsciousmusings/2017/04/12/machine-learning-algorithm-use/

The Machine Learning Primer (sas.com)

https://www.sas.com/content/dam/SAS/en_us/doc/whitepaper1/machine-learning-primer-108796.pdf

Machine Learning Tutorial for Beginners (kaggle.com/kanncaa1)

https://www.kaggle.com/kanncaa1/machine-learning-tutorial-for-beginners

激活和损失函数

Sigmoid neurons (neuralnetworksanddeeplearning.com)

http://neuralnetworksanddeeplearning.com/chap1.html#sigmoid_neurons

What is the role of the activation function in a neural network? (quora.com)

https://www.quora.com/What-is-the-role-of-the-activation-function-in-a-neural-network

Comprehensive list of activation functions in neural networks with pros/cons(stats.stackexchange.com)

https://stats.stackexchange.com/questions/115258/comprehensive-list-of-activation-functions-in-neural-networks-with-pros-cons

Activation functions and it’s types-Which is better? (medium.com)

https://medium.com/towards-data-science/activation-functions-and-its-types-which-is-better-a9a5310cc8f

Making Sense of Logarithmic Loss (exegetic.biz)

http://www.exegetic.biz/blog/2015/12/making-sense-logarithmic-loss/

Loss Functions (Stanford CS231n)

http://cs231n.github.io/neural-networks-2/#losses

L1 vs. L2 Loss function (rishy.github.io)

http://rishy.github.io/ml/2015/07/28/l1-vs-l2-loss/

The cross-entropy cost function (neuralnetworksanddeeplearning.com)

http://neuralnetworksanddeeplearning.com/chap3.html#the_cross-entropy_cost_function

偏差

Role of Bias in Neural Networks (stackoverflow.com)

https://stackoverflow.com/questions/2480650/role-of-bias-in-neural-networks/2499936#2499936

Bias Nodes in Neural Networks(makeyourownneuralnetwork.blogspot.com)

http://makeyourownneuralnetwork.blogspot.com/2016/06/bias-nodes-in-neural-networks.html

What is bias in artificial neural network? (quora.com)

https://www.quora.com/What-is-bias-in-artificial-neural-network

感知

Perceptrons (neuralnetworksanddeeplearning.com)

http://neuralnetworksanddeeplearning.com/chap1.html#perceptrons

The Perception (natureofcode.com)

https://natureofcode.com/book/chapter-10-neural-networks/#chapter10_figure3

Single-layer Neural Networks (Perceptrons) (dcu.ie)

http://computing.dcu.ie/~humphrys/Notes/Neural/single.neural.html

From Perceptrons to Deep Networks (toptal.com)

https://www.toptal.com/machine-learning/an-introduction-to-deep-learning-from-perceptrons-to-deep-networks

回归

Introduction to linear regression analysis (duke.edu)

http://people.duke.edu/~rnau/regintro.htm

Linear Regression (ufldl.stanford.edu)

http://ufldl.stanford.edu/tutorial/supervised/LinearRegression/

Linear Regression (readthedocs.io)

http://ml-cheatsheet.readthedocs.io/en/latest/linear_regression.html

Logistic Regression (readthedocs.io)

https://ml-cheatsheet.readthedocs.io/en/latest/logistic_regression.html

Simple Linear Regression Tutorial for Machine Learning(machinelearningmastery.com)

http://machinelearningmastery.com/simple-linear-regression-tutorial-for-machine-learning/

Logistic Regression Tutorial for Machine Learning(machinelearningmastery.com)

https://machinelearningmastery.com/logistic-regression-tutorial-for-machine-learning/

Softmax Regression (ufldl.stanford.edu)

http://ufldl.stanford.edu/tutorial/supervised/SoftmaxRegression/

梯度下降

Learning with gradient descent (neuralnetworksanddeeplearning.com)

http://neuralnetworksanddeeplearning.com/chap1.html#learning_with_gradient_descent

Gradient Descent (iamtrask.github.io)

http://iamtrask.github.io/2015/07/27/python-network-part2/

How to understand Gradient Descent algorithm (kdnuggets.com)

http://www.kdnuggets.com/2017/04/simple-understand-gradient-descent-algorithm.html

An overview of gradient descent optimization algorithms(sebastianruder.com)

http://sebastianruder.com/optimizing-gradient-descent/

Optimization: Stochastic Gradient Descent (Stanford CS231n)

http://cs231n.github.io/optimization-1/

生成学习

Generative Learning Algorithms (Stanford CS229)

http://cs229.stanford.edu/notes/cs229-notes2.pdf

A practical explanation of a Naive Bayes classifier (monkeylearn.com)

https://monkeylearn.com/blog/practical-explanation-naive-bayes-classifier/

支持向量机

An introduction to Support Vector Machines (SVM) (monkeylearn.com)

https://monkeylearn.com/blog/introduction-to-support-vector-machines-svm/

Support Vector Machines (Stanford CS229)

http://cs229.stanford.edu/notes/cs229-notes3.pdf

Linear classification: Support Vector Machine, Softmax (Stanford 231n)

http://cs231n.github.io/linear-classify/

深度学习

A Guide to Deep Learning by YN² (yerevann.com)

http://yerevann.com/a-guide-to-deep-learning/

Deep Learning Papers Reading Roadmap (github.com/floodsung)

https://github.com/floodsung/Deep-Learning-Papers-Reading-Roadmap

Deep Learning in a Nutshell (nikhilbuduma.com)

http://nikhilbuduma.com/2014/12/29/deep-learning-in-a-nutshell/

A Tutorial on Deep Learning (Quoc V. Le)

http://ai.stanford.edu/~quocle/tutorial1.pdf

What is Deep Learning? (machinelearningmastery.com)

https://machinelearningmastery.com/what-is-deep-learning/

What’s the Difference Between Artificial Intelligence, Machine Learning, and Deep Learning? (nvidia.com)

https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/

Deep Learning — The Straight Dope (gluon.mxnet.io)

https://gluon.mxnet.io/

优化和降维

Seven Techniques for Data Dimensionality Reduction (knime.org)

https://www.knime.org/blog/seven-techniques-for-data-dimensionality-reduction

Principal components analysis (Stanford CS229)

http://cs229.stanford.edu/notes/cs229-notes10.pdf

Dropout: A simple way to improve neural networks (Hinton @ NIPS 2012)

http://cs229.stanford.edu/notes/cs229-notes10.pdf

How to train your Deep Neural Network (rishy.github.io)

http://rishy.github.io/ml/2017/01/05/how-to-train-your-dnn/

长短期记忆(LSTM

A Gentle Introduction to Long Short-Term Memory Networks by the Experts(machinelearningmastery.com)

https://machinelearningmastery.com/gentle-introduction-long-short-term-memory-networks-experts/

Understanding LSTM Networks (colah.github.io)

http://colah.github.io/posts/2015-08-Understanding-LSTMs/

Exploring LSTMs (echen.me)

http://blog.echen.me/2017/05/30/exploring-lstms/

Anyone Can Learn To Code an LSTM-RNN in Python (iamtrask.github.io)

http://iamtrask.github.io/2015/11/15/anyone-can-code-lstm/

卷积神经网络

Introducing convolutional networks (neuralnetworksanddeeplearning.com)

http://neuralnetworksanddeeplearning.com/chap6.html#introducing_convolutional_networks

Deep Learning and Convolutional Neural Networks(medium.com/@ageitgey)

https://medium.com/@ageitgey/machine-learning-is-fun-part-3-deep-learning-and-convolutional-neural-networks-f40359318721

Conv Nets: A Modular Perspective (colah.github.io)

http://colah.github.io/posts/2014-07-Conv-Nets-Modular/

Understanding Convolutions (colah.github.io)

http://colah.github.io/posts/2014-07-Understanding-Convolutions/

递归神经网络

Recurrent Neural Networks Tutorial (wildml.com)

http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/

Attention and Augmented Recurrent Neural Networks (distill.pub)

http://distill.pub/2016/augmented-rnns/

The Unreasonable Effectiveness of Recurrent Neural Networks(karpathy.github.io)

http://karpathy.github.io/2015/05/21/rnn-effectiveness/

A Deep Dive into Recurrent Neural Nets (nikhilbuduma.com)

http://nikhilbuduma.com/2015/01/11/a-deep-dive-into-recurrent-neural-networks/

强化学习

Simple Beginner’s guide to Reinforcement Learning & its implementation(analyticsvidhya.com)

https://www.analyticsvidhya.com/blog/2017/01/introduction-to-reinforcement-learning-implementation/

A Tutorial for Reinforcement Learning (mst.edu)

https://web.mst.edu/~gosavia/tutorial.pdf

Learning Reinforcement Learning (wildml.com)

http://www.wildml.com/2016/10/learning-reinforcement-learning/

Deep Reinforcement Learning: Pong from Pixels (karpathy.github.io)

http://karpathy.github.io/2016/05/31/rl/

生成对抗网络(GANs)

Adversarial Machine Learning (aaai18adversarial.github.io)

https://aaai18adversarial.github.io/slides/AML.pptx

What’s a Generative Adversarial Network? (nvidia.com)

https://blogs.nvidia.com/blog/2017/05/17/generative-adversarial-network/

Abusing Generative Adversarial Networks to Make 8-bit Pixel Art(medium.com/@ageitgey)

https://medium.com/@ageitgey/abusing-generative-adversarial-networks-to-make-8-bit-pixel-art-e45d9b96cee7

An introduction to Generative Adversarial Networks (with code in TensorFlow) (aylien.com)

http://blog.aylien.com/introduction-generative-adversarial-networks-code-tensorflow/

Generative Adversarial Networks for Beginners (oreilly.com)

https://www.oreilly.com/learning/generative-adversarial-networks-for-beginners

多任务学习

An Overview of Multi-Task Learning in Deep Neural Networks(sebastianruder.com)

http://sebastianruder.com/multi-task/index.html

自然语言处理

Natural Language Processing is Fun! (medium.com/@ageitgey)

https://medium.com/@ageitgey/natural-language-processing-is-fun-9a0bff37854e

A Primer on Neural Network Models for Natural Language Processing(Yoav Goldberg)

http://u.cs.biu.ac.il/~yogo/nnlp.pdf

The Definitive Guide to Natural Language Processing (monkeylearn.com)

https://monkeylearn.com/blog/the-definitive-guide-to-natural-language-processing/

Introduction to Natural Language Processing (algorithmia.com)

https://blog.algorithmia.com/introduction-natural-language-processing-nlp/

Natural Language Processing Tutorial (vikparuchuri.com)

http://www.vikparuchuri.com/blog/natural-language-processing-tutorial/

Natural Language Processing (almost) from Scratch (arxiv.org)

https://arxiv.org/pdf/1103.0398.pdf

深度学习自然语言处理

Deep Learning applied to NLP (arxiv.org)

https://arxiv.org/pdf/1703.03091.pdf

Deep Learning for NLP (without Magic) (Richard Socher)

https://nlp.stanford.edu/courses/NAACL2013/NAACL2013-Socher-Manning-DeepLearning.pdf

Understanding Convolutional Neural Networks for NLP (wildml.com)

http://www.wildml.com/2015/11/understanding-convolutional-neural-networks-for-nlp/

Deep Learning, NLP, and Representations (colah.github.io)

http://colah.github.io/posts/2014-07-NLP-RNNs-Representations/

Embed, encode, attend, predict: The new deep learning formula for state-of-the-art NLP models (explosion.ai)

https://explosion.ai/blog/deep-learning-formula-nlp

Understanding Natural Language with Deep Neural Networks Using Torch(nvidia.com)

https://devblogs.nvidia.com/parallelforall/understanding-natural-language-deep-neural-networks-using-torch/

Deep Learning for NLP with Pytorch (pytorich.org)                                            

http://pytorch.org/tutorials/beginner/deep_learning_nlp_tutorial.html

词向量

Bag of Words Meets Bags of Popcorn (kaggle.com)

https://www.kaggle.com/c/word2vec-nlp-tutorial

On word embeddings Part I, Part II, Part III (sebastianruder.com)

http://sebastianruder.com/word-embeddings-1/index.html

http://sebastianruder.com/word-embeddings-softmax/index.html

http://sebastianruder.com/secret-word2vec/index.html

The amazing power of word vectors (acolyer.org)

https://blog.acolyer.org/2016/04/21/the-amazing-power-of-word-vectors/

word2vec Parameter Learning Explained (arxiv.org)

https://arxiv.org/pdf/1411.2738.pdf

Word2Vec Tutorial — The Skip-Gram Model, Negative Sampling(mccormickml.com)

http://mccormickml.com/2016/04/19/word2vec-tutorial-the-skip-gram-model/

http://mccormickml.com/2017/01/11/word2vec-tutorial-part-2-negative-sampling/

编码器-解码器

Attention and Memory in Deep Learning and NLP (wildml.com)

http://www.wildml.com/2016/01/attention-and-memory-in-deep-learning-and-nlp/

Sequence to Sequence Models (tensorflow.org)

https://www.tensorflow.org/tutorials/seq2seq

Sequence to Sequence Learning with Neural Networks (NIPS 2014)

https://papers.nips.cc/paper/5346-sequence-to-sequence-learning-with-neural-networks.pdf

Machine Learning is Fun Part 5: Language Translation with Deep Learning and the Magic of Sequences (medium.com/@ageitgey)

https://medium.com/@ageitgey/machine-learning-is-fun-part-5-language-translation-with-deep-learning-and-the-magic-of-sequences-2ace0acca0aa

tf-seq2seq (google.github.io)

https://google.github.io/seq2seq/

Python

Machine Learning Crash Course (google.com)

https://developers.google.com/machine-learning/crash-course/

Awesome Machine Learning (github.com/josephmisiti)

https://github.com/josephmisiti/awesome-machine-learning#python

7 Steps to Mastering Machine Learning With Python (kdnuggets.com)

http://www.kdnuggets.com/2015/11/seven-steps-machine-learning-python.html

An example machine learning notebook (nbviewer.jupyter.org)

http://nbviewer.jupyter.org/github/rhiever/Data-Analysis-and-Machine-Learning-Projects/blob/master/example-data-science-notebook/Example%20Machine%20Learning%20Notebook.ipynb

Machine Learning with Python (tutorialspoint.com)

https://www.tutorialspoint.com/machine_learning_with_python/machine_learning_with_python_quick_guide.htm

范例

How To Implement The Perceptron Algorithm From Scratch In Python(machinelearningmastery.com)

http://machinelearningmastery.com/implement-perceptron-algorithm-scratch-python/

Implementing a Neural Network from Scratch in Python (wildml.com)

http://www.wildml.com/2015/09/implementing-a-neural-network-from-scratch/

A Neural Network in 11 lines of Python (iamtrask.github.io)

http://iamtrask.github.io/2015/07/12/basic-python-network/

Implementing Your Own k-Nearest Neighbour Algorithm Using Python(kdnuggets.com)

http://www.kdnuggets.com/2016/01/implementing-your-own-knn-using-python.html

ML from Scatch (github.com/eriklindernoren)

https://github.com/eriklindernoren/ML-From-Scratch

Python Machine Learning (2nd Ed.) Code Repository (github.com/rasbt)

https://github.com/rasbt/python-machine-learning-book-2nd-edition

Scipy and numpy

Scipy Lecture Notes (scipy-lectures.org)

http://www.scipy-lectures.org/

Python Numpy Tutorial (Stanford CS231n)

http://cs231n.github.io/python-numpy-tutorial/

An introduction to Numpy and Scipy (UCSB CHE210D)

https://engineering.ucsb.edu/~shell/che210d/numpy.pdf

A Crash Course in Python for Scientists (nbviewer.jupyter.org)

http://nbviewer.jupyter.org/gist/rpmuller/5920182#ii.-numpy-and-scipy

scikit-learn

PyCon scikit-learn Tutorial Index (nbviewer.jupyter.org)

http://nbviewer.jupyter.org/github/jakevdp/sklearn_pycon2015/blob/master/notebooks/Index.ipynb

scikit-learn Classification Algorithms (github.com/mmmayo13)

https://github.com/mmmayo13/scikit-learn-classifiers/blob/master/sklearn-classifiers-tutorial.ipynb

scikit-learn Tutorials (scikit-learn.org)

http://scikit-learn.org/stable/tutorial/index.html

Abridged scikit-learn Tutorials (github.com/mmmayo13)

https://github.com/mmmayo13/scikit-learn-beginners-tutorials

Tensorflow

Tensorflow Tutorials (tensorflow.org)

https://www.tensorflow.org/tutorials/

Introduction to TensorFlow — CPU vs GPU (medium.com/@erikhallstrm)

https://medium.com/@erikhallstrm/hello-world-tensorflow-649b15aed18c

TensorFlow: A primer (metaflow.fr)

https://blog.metaflow.fr/tensorflow-a-primer-4b3fa0978be3

RNNs in Tensorflow (wildml.com)

http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/

Implementing a CNN for Text Classification in TensorFlow (wildml.com)

http://www.wildml.com/2015/12/implementing-a-cnn-for-text-classification-in-tensorflow/

How to Run Text Summarization with TensorFlow (surmenok.com)

http://pavel.surmenok.com/2016/10/15/how-to-run-text-summarization-with-tensorflow/

PyTorch

PyTorch Tutorials (pytorch.org)

http://pytorch.org/tutorials/

A Gentle Intro to PyTorch (gaurav.im)

http://blog.gaurav.im/2017/04/24/a-gentle-intro-to-pytorch/

Tutorial: Deep Learning in PyTorch (iamtrask.github.io)

https://iamtrask.github.io/2017/01/15/pytorch-tutorial/

PyTorch Examples (github.com/jcjohnson)

https://github.com/jcjohnson/pytorch-examples

PyTorch Tutorial (github.com/MorvanZhou)

https://github.com/MorvanZhou/PyTorch-Tutorial

PyTorch Tutorial for Deep Learning Researchers (github.com/yunjey)

https://github.com/yunjey/pytorch-tutorial

数学

Math for Machine Learning (ucsc.edu)

https://people.ucsc.edu/~praman1/static/pub/math-for-ml.pdf

Math for Machine Learning (UMIACS CMSC422)

http://www.umiacs.umd.edu/~hal/courses/2013S_ML/math4ml.pdf

线性代数

An Intuitive Guide to Linear Algebra (betterexplained.com)

https://betterexplained.com/articles/linear-algebra-guide/

A Programmer’s Intuition for Matrix Multiplication (betterexplained.com)

https://betterexplained.com/articles/matrix-multiplication/

Understanding the Cross Product (betterexplained.com)

https://betterexplained.com/articles/cross-product/

Understanding the Dot Product (betterexplained.com)

https://betterexplained.com/articles/vector-calculus-understanding-the-dot-product/

Linear Algebra for Machine Learning (U. of Buffalo CSE574)

http://www.cedar.buffalo.edu/~srihari/CSE574/Chap1/LinearAlgebra.pdf

Linear algebra cheat sheet for deep learning (medium.com)

https://medium.com/towards-data-science/linear-algebra-cheat-sheet-for-deep-learning-cd67aba4526c

Linear Algebra Review and Reference (Stanford CS229)

http://cs229.stanford.edu/section/cs229-linalg.pdf

概率

Understanding Bayes Theorem With Ratios (betterexplained.com)

https://betterexplained.com/articles/understanding-bayes-theorem-with-ratios/

Review of Probability Theory (Stanford CS229)

http://cs229.stanford.edu/section/cs229-prob.pdf

Probability Theory Review for Machine Learning (Stanford CS229)

https://see.stanford.edu/materials/aimlcs229/cs229-prob.pdf

Probability Theory (U. of Buffalo CSE574)

http://www.cedar.buffalo.edu/~srihari/CSE574/Chap1/Probability-Theory.pdf

Probability Theory for Machine Learning (U. of Toronto CSC411)

http://www.cs.toronto.edu/~urtasun/courses/CSC411_Fall16/tutorial1.pdf

微积分

How To Understand Derivatives: The Quotient Rule, Exponents, and Logarithms (betterexplained.com)

https://betterexplained.com/articles/how-to-understand-derivatives-the-quotient-rule-exponents-and-logarithms/

How To Understand Derivatives: The Product, Power & Chain Rules(betterexplained.com)

https://betterexplained.com/articles/derivatives-product-power-chain/

Vector Calculus: Understanding the Gradient (betterexplained.com)

https://betterexplained.com/articles/vector-calculus-understanding-the-gradient/

Differential Calculus (Stanford CS224n)

http://web.stanford.edu/class/cs224n/lecture_notes/cs224n-2017-review-differential-calculus.pdf

Calculus Overview (readthedocs.io)

http://ml-cheatsheet.readthedocs.io/en/latest/calculus.html

大数据文摘
大数据文摘

秉承“普及数据思维,传播数据文化,助⼒产业发展”的企业⽂化,我们专注于数据领域的资讯、案例、技术,形成了“媒体+教育+⼈才服务”的良性⽣态,致⼒于打造精准数据科学社区。

入门机器学习NLPPython
25
相关数据
理查德·索切人物

Richard Socher(理查德·索赫尔)是Salesforce的首席科学家。 在此之前,他是斯坦福大学计算机科学系的兼职教授,也是2016年被Salesforce收购的MetaMind的创始人兼首席执行官/首席技术官。研究兴趣:深度学习、自然语言处理和计算机视觉。

深度学习技术

深度学习(deep learning)是机器学习的分支,是一种试图使用包含复杂结构或由多重非线性变换构成的多个处理层对数据进行高层抽象的算法。 深度学习是机器学习中一种基于对数据进行表征学习的算法,至今已有数种深度学习框架,如卷积神经网络和深度置信网络和递归神经网络等已被应用在计算机视觉、语音识别、自然语言处理、音频识别与生物信息学等领域并获取了极好的效果。

激活函数技术

在 计算网络中, 一个节点的激活函数定义了该节点在给定的输入或输入的集合下的输出。标准的计算机芯片电路可以看作是根据输入得到"开"(1)或"关"(0)输出的数字网络激活函数。这与神经网络中的线性感知机的行为类似。 一种函数(例如 ReLU 或 S 型函数),用于对上一层的所有输入求加权和,然后生成一个输出值(通常为非线性值),并将其传递给下一层。

机器学习技术

机器学习是人工智能的一个分支,是一门多领域交叉学科,涉及概率论、统计学、逼近论、凸分析、计算复杂性理论等多门学科。机器学习理论主要是设计和分析一些让计算机可以自动“学习”的算法。因为学习算法中涉及了大量的统计学理论,机器学习与推断统计学联系尤为密切,也被称为统计学习理论。算法设计方面,机器学习理论关注可以实现的,行之有效的学习算法。

Dropout技术

神经网络训练中防止过拟合的一种技术

感知技术

知觉或感知是外界刺激作用于感官时,脑对外界的整体的看法和理解,为我们对外界的感官信息进行组织和解释。在认知科学中,也可看作一组程序,包括获取信息、理解信息、筛选信息、组织信息。与感觉不同,知觉反映的是由对象的各样属性及关系构成的整体。

人工智能技术

在学术研究领域,人工智能通常指能够感知周围环境并采取行动以实现最优的可能结果的智能体(intelligent agent)

损失函数技术

在数学优化,统计学,计量经济学,决策理论,机器学习和计算神经科学等领域,损失函数或成本函数是将一或多个变量的一个事件或值映射为可以直观地表示某种与之相关“成本”的实数的函数。

TensorFlow技术

TensorFlow是一个开源软件库,用于各种感知和语言理解任务的机器学习。目前被50个团队用于研究和生产许多Google商业产品,如语音识别、Gmail、Google 相册和搜索,其中许多产品曾使用过其前任软件DistBelief。

张量技术

张量是一个可用来表示在一些矢量、标量和其他张量之间的线性关系的多线性函数,这些线性关系的基本例子有内积、外积、线性映射以及笛卡儿积。其坐标在 维空间内,有 个分量的一种量,其中每个分量都是坐标的函数,而在坐标变换时,这些分量也依照某些规则作线性变换。称为该张量的秩或阶(与矩阵的秩和阶均无关系)。 在数学里,张量是一种几何实体,或者说广义上的“数量”。张量概念包括标量、矢量和线性算子。张量可以用坐标系统来表达,记作标量的数组,但它是定义为“不依赖于参照系的选择的”。张量在物理和工程学中很重要。例如在扩散张量成像中,表达器官对于水的在各个方向的微分透性的张量可以用来产生大脑的扫描图。工程上最重要的例子可能就是应力张量和应变张量了,它们都是二阶张量,对于一般线性材料他们之间的关系由一个四阶弹性张量来决定。

神经网络技术

(人工)神经网络是一种起源于 20 世纪 50 年代的监督式机器学习模型,那时候研究者构想了「感知器(perceptron)」的想法。这一领域的研究者通常被称为「联结主义者(Connectionist)」,因为这种模型模拟了人脑的功能。神经网络模型通常是通过反向传播算法应用梯度下降训练的。目前神经网络有两大主要类型,它们都是前馈神经网络:卷积神经网络(CNN)和循环神经网络(RNN),其中 RNN 又包含长短期记忆(LSTM)、门控循环单元(GRU)等等。深度学习是一种主要应用于神经网络帮助其取得更好结果的技术。尽管神经网络主要用于监督学习,但也有一些为无监督学习设计的变体,比如自动编码器和生成对抗网络(GAN)。

梯度下降技术

梯度下降是用于查找函数最小值的一阶迭代优化算法。 要使用梯度下降找到函数的局部最小值,可以采用与当前点的函数梯度(或近似梯度)的负值成比例的步骤。 如果采取的步骤与梯度的正值成比例,则接近该函数的局部最大值,被称为梯度上升。

卷积神经网络技术

卷积神经网路(Convolutional Neural Network, CNN)是一种前馈神经网络,它的人工神经元可以响应一部分覆盖范围内的周围单元,对于大型图像处理有出色表现。卷积神经网路由一个或多个卷积层和顶端的全连通层(对应经典的神经网路)组成,同时也包括关联权重和池化层(pooling layer)。这一结构使得卷积神经网路能够利用输入数据的二维结构。与其他深度学习结构相比,卷积神经网路在图像和语音识别方面能够给出更好的结果。这一模型也可以使用反向传播算法进行训练。相比较其他深度、前馈神经网路,卷积神经网路需要考量的参数更少,使之成为一种颇具吸引力的深度学习结构。 卷积网络是一种专门用于处理具有已知的、网格状拓扑的数据的神经网络。例如时间序列数据,它可以被认为是以一定时间间隔采样的一维网格,又如图像数据,其可以被认为是二维像素网格。

支持向量机技术

在机器学习中,支持向量机是在分类与回归分析中分析数据的监督式学习模型与相关的学习算法。给定一组训练实例,每个训练实例被标记为属于两个类别中的一个或另一个,SVM训练算法创建一个将新的实例分配给两个类别之一的模型,使其成为非概率二元线性分类器。SVM模型是将实例表示为空间中的点,这样映射就使得单独类别的实例被尽可能宽的明显的间隔分开。然后,将新的实例映射到同一空间,并基于它们落在间隔的哪一侧来预测所属类别。

神经元技术

(人工)神经元是一个类比于生物神经元的数学计算模型,是神经网络的基本组成单元。 对于生物神经网络,每个神经元与其他神经元相连,当它“兴奋”时会向相连的神经元发送化学物质,从而改变这些神经元的电位;神经元的“兴奋”由其电位决定,当它的电位超过一个“阈值”(threshold)便会被激活,亦即“兴奋”。 目前最常见的神经元模型是基于1943年 Warren McCulloch 和 Walter Pitts提出的“M-P 神经元模型”。 在这个模型中,神经元通过带权重的连接接处理来自n个其他神经元的输入信号,其总输入值将与神经元的阈值进行比较,最后通过“激活函数”(activation function)产生神经元的输出。

降维技术

降维算法是将 p+1 个系数的问题简化为 M+1 个系数的问题,其中 M<p。算法执行包括计算变量的 M 个不同线性组合或投射(projection)。然后这 M 个投射作为预测器通过最小二乘法拟合一个线性回归模型。两个主要的方法是主成分回归(principal component regression)和偏最小二乘法(partial least squares)。

word2vec技术

Word2vec,为一群用来产生词向量的相关模型。这些模型为浅而双层的神经网络,用来训练以重新建构语言学之词文本。网络以词表现,并且需猜测相邻位置的输入词,在word2vec中词袋模型假设下,词的顺序是不重要的。 训练完成之后,word2vec模型可用来映射每个词到一个向量,可用来表示词对词之间的关系。该向量为神经网络之隐藏层。 Word2vec依赖skip-grams或连续词袋(CBOW)来建立神经词嵌入。Word2vec为托马斯·米科洛夫(Tomas Mikolov)在Google带领的研究团队创造。该算法渐渐被其他人所分析和解释。

自然语言处理技术

自然语言处理(英语:natural language processing,缩写作 NLP)是人工智能和语言学领域的分支学科。此领域探讨如何处理及运用自然语言;自然语言认知则是指让电脑“懂”人类的语言。自然语言生成系统把计算机数据转化为自然语言。自然语言理解系统把自然语言转化为计算机程序更易于处理的形式。

生成对抗网络技术

生成对抗网络是一种无监督学习方法,是一种通过用对抗网络来训练生成模型的架构。它由两个网络组成:用来拟合数据分布的生成网络G,和用来判断输入是否“真实”的判别网络D。在训练过程中,生成网络-G通过接受一个随机的噪声来尽量模仿训练集中的真实图片去“欺骗”D,而D则尽可能的分辨真实数据和生成网络的输出,从而形成两个网络的博弈过程。理想的情况下,博弈的结果会得到一个可以“以假乱真”的生成模型。

多任务学习技术

长短期记忆网络技术

长短期记忆(Long Short-Term Memory) 是具有长期记忆能力的一种时间递归神经网络(Recurrent Neural Network)。 其网络结构含有一个或多个具有可遗忘和记忆功能的单元组成。它在1997年被提出用于解决传统RNN(Recurrent Neural Network) 的随时间反向传播中权重消失的问题(vanishing gradient problem over backpropagation-through-time),重要组成部分包括Forget Gate, Input Gate, 和 Output Gate, 分别负责决定当前输入是否被采纳,是否被长期记忆以及决定在记忆中的输入是否在当前被输出。Gated Recurrent Unit 是 LSTM 众多版本中典型的一个。因为它具有记忆性的功能,LSTM经常被用在具有时间序列特性的数据和场景中。

强化学习技术

强化学习是一种试错方法,其目标是让软件智能体在特定环境中能够采取回报最大化的行为。强化学习在马尔可夫决策过程环境中主要使用的技术是动态规划(Dynamic Programming)。流行的强化学习方法包括自适应动态规划(ADP)、时间差分(TD)学习、状态-动作-回报-状态-动作(SARSA)算法、Q 学习、深度强化学习(DQN);其应用包括下棋类游戏、机器人控制和工作调度等。

感知器技术

感知器是Frank Rosenblatt在1957年就职于Cornell航空实验室时所发明的一种人工神经网络。它可以被视为一种最简单形式的前馈神经网络,是一种二元线性分类器。 Frank Rosenblatt给出了相应的感知机学习算法,常用的有感知机学习、最小二乘法和梯度下降法。

暂无评论
暂无评论~