Read about stanford university machine learning coursera, The latest news, videos, and discussion topics about stanford university machine learning coursera from alibabacloud.com
This is what we have learned (except decision tree)Here is a typical decision tree algorithm, with four places to choose from:Then introduced a cart algorithm: By decision Stump divided into two categories, the criterion for measuring subtree is that the data are divided into two categories, the purity of these two types of data (purifying).The following is a measure of purity:Finally, when to stop:Decision tree may be overfitting, reducing the number of Ein and leaves (indicating the complexity
cost function least.The algorithm is:After derivation, get:Note: Although the resulting gradient descent algorithm appears to be the same as the gradient descent algorithm for linear regression, the hypothetical function here differs from the linear regression, so it is actually different. In addition, it is still necessary to perform feature scaling before applying the gradient descent algorithm.In addition, there are some alternatives to the gradient descent algorithm:In addition to the gradi
In this section, a linear model is introduced, and several linear models are compared, and the linear regression and the logistic regression are used for classification by the conversion error function.More important is this diagram, which explains why you can use linear regression or a logistic regression to replace linear classificationThen the stochastic gradient descent method is introduced, which is an improvement to the gradient descent method, which greatly improves the efficiency.Finally
This section is about the nuclear svm,andrew Ng's handout, which is also well-spoken.The first is kernel trick, which uses nuclear techniques to simplify the calculation of low-dimensional features by mapping high-dimensional features. The handout also speaks of the determination of the kernel function, that is, what function K can use kernel trick.In addition, the kernel function can measure the similarity of two features, the greater the value, the more similar.Next is the polynomial Kernel, w
-Polynomial regressionSince linear regression does not apply to all data, sometimes we need to use curves to fit our data, for example, with two-times polynomial:Or three-time polynomial:Usually we need to look at the data before deciding what model to try to fit.After that, we can make:The two-time polynomial is then converted to a linear regression model.It is worth noting that if we adopt a polynomial regression model, feature scaling is necessary before the gradient descent algorithm is run.
the transpose of the Matrix.-Gradient descent for multiple variablesSimilar to univariate/feature linear regression, in multivariable/feature linear regression, we will also define a cost function, namely:Our goal is the same as the problem in univariate/characteristic linear regression, which is to find out the combination of parameters that make the cost function least.Therefore, the multivariable/linear regression gradient descent algorithm is:ThatAfter the derivative number can be obtained:
-Normal equationSo far, the gradient descent algorithm has been used in linear regression problems, but for some linear regression problems, the normal equation method is a better solution.The normal equation is solved by solving the following equations to find the parameters that make the cost function least:Assuming our training set feature matrix is x, our training set results are vector y, then the normal equation is used to solve the vector:The following table shows the data as an example:T
17.1 Study of large data sets17.2 Random Gradient Descent method17.3 Miniature Batch Gradient descent17.4 Stochastic gradient descent convergence17.5 Online Learning17.6 mapping simplification and data parallelism
17.1 Learning from large data sets
17.2random Gradient Descent method
17.3miniature Batch gradient descent
17.4stochastic gradient descent convergence
17.5Online Learning
calculates the accuracy of the entire system at this time:
As shown in, text recognition consists of four parts. Now we can find the system accuracy after optimization for each part. The question is, how can we improve the accuracy of the entire system? We can see from the table that, if we have optimized the text moderation part, the accuracy will be72%Add89%If we optimize the character segmentation, the accuracy is only from89%To90%If character recognition is optimized90%To100%In contr
11.1 What to do first11.2 Error AnalysisError measurement for class 11.3 skew11.4 The tradeoff between recall and precision11.5 Machine-Learning data
11.1 what to do firstIn the next video, I'll talk about the design of the machine learning system. These videos will talk about the major problems you will encounte
An introductory tutorial on machine learning with a higher degree of identity, by Andrew Ng of Stanford. NetEase public class with Chinese and English subtitles teaching video resources (http://open.163.com/special/opencourse/ machinelearning.html), handout stamp here: http://cs229.stanford.edu/materials.htmlThere are a variety of similar course
It is decided that machine learning is under system learning, and Stanford courseware is the main line.
Notes1 is part of the http://www.stanford.edu/class/cs229/notes/cs229-notes1.pdf about Regression 1. Linear Regression
For example, if the House Price is predicted and the data cannot be found on the Internet, use
invoking the example in MATLAB above, we can define the cost function of the logistic regression as follows:In the figure, Jval represents the cost function expression, where the last item is the penalty for the parameter θ; The following is a gradient of the derivation of each θj, where θ0 is not in the penalty, so gradient is not changed, and Θ1~θn has one more (λ/m) *θj respectively;At this point, regularization can solve the linear and logistic overfitting regression problem ~
Since the end of last year to learn Andrew Ng's machine learning public class, in accordance with its courseware to try to achieve some of the algorithm to deepen understanding, but in this process encountered some problems, or for the implementation of the program, or to understand the algorithm. So prepare to organize this course and document your understanding, either right or wrong, to discuss together.
function and the derivation of each parameter when using it. we implement the costfunction ourselves and pass in the response parameter. We can return the following two values at a time:
For example, call the fminunc () function and use @ to input the pointer to the costfunction function. For the initialized Theta, you can also add options (gradobj = on indicates "Open the gradient target parameter ", that is, we will provide gradient parameters for this function ):
6.7 multi-category classifi
It should be this time last year, I started to get into the knowledge of machine learning, then the introductory book is "Introduction to data mining." Swallowed read the various well-known classifiers: Decision Tree, naive Bayesian, SVM, neural network, random forest and so on; In addition, more serious review of statistics, learning the linear regression, but a
is more than one, the Newton method iterates over the rule:Newton's method usually has a faster convergence rate than the batch gradient, and it takes a much smaller number of iterations to get close to the minimum value. However, when the parameters of the model are many (n), the computational cost of the Hessian matrix will be large, resulting in a slower convergence rate, but when the number of arguments is not long, the Newton method is usually much faster than the gradient descent.Summariz
based on the minimum mean variance. The closer to the predicted point, the heavier the weight, which is to use the points near the check to give higher weights. The most common is the Gaussian nucleus. The weights corresponding to the Gaussian nuclei are as follows:In (Formula 2), the only thing we need to make sure is that it's a user-specified parameter that determines how much weight is given to nearby points.Therefore, as shown in (Equation 3), local weighted linear regression is a non-para
symmetric semi-definite matrixin the case where the data is non-linear:called L1 norm soft margin SVM. is a convex optimization problem. It allows an interval of less than 1, which allows for the categorization of errors. SMO algorithm:coordinate ascent algorithm:This algorithm has more iterations, but at some point the inner loop will be very fast if a parameter in W (A1,,, am) is very small at the cost of finding the optimal value. SMO:If only one α is solved as SVM, the other α is fixed. obt
The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion;
products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the
content of the page makes you feel confusing, please write us an email, we will handle the problem
within 5 days after receiving your email.
If you find any instances of plagiarism from the community, please send an email to:
info-contact@alibabacloud.com
and provide relevant evidence. A staff member will contact you within 5 working days.