List of machine learning resources. Part 1
Earlier we talked about the development of a quantum communication system and how advanced programmers are trained from simple students. Today we decide once again ( 1 , 2 ) to look towards the topic of machine learning and provide an adapted ( source ) selection of useful materials discussed on the Stack Overflow and the Stack Exchange.
Genetic algorithms
- SO: genetic algorithms and artificial neural networks;
- SO: genetic algorithms and genetic programming;
Classification
- SO: when to choose a classifier;
Linear regression
- SE: what is the same variance of linear regression model errors;
- SE: what is the difference between linear regression of y for x and linear regression of x for y;
- SE: interpretation of plot.lm () in R;
- SE: interpretation of quantile-quantile graphics;
- SE: Interpretation of graphs Residuals vs Fitted;
- SE: how to handle abnormal values;
Logistic regression
- SE: obtaining predicted Y values;
- SE: residuals in logistic regression;
- SE: Differences between logistic regression and probit regression;
- SE: pseudo R-squared and logistic regression;
- SE: how to calculate a pseudo R-square;
Model validation using resampling
- SE: split dataset in R;
- SE: score c splitting the sample in R;
- SE: learning with a full set of data after cross-checking;
- SE: best cross validation method;
- SE: estimation of variance in cross-checking by k-blocks;
- SE: whether cross-validation can replace the control sample;
- SE: select the number of blocks for cross-checking by k-blocks;
- SE: cross validation for compositional learning;
- SE: how cross-validation solves the problem of retraining;
- SE: why bootstrap works;
- SE: statistical bootstrap for model selection and evaluation;
- SE: using cross-validation and boost trap for estimating prediction errors;
- SE: what to use to assess the effectiveness of the classification - cross-checking or bootstrap;
- SO: cross-checking by k-blocks in R;
Deep learning
- SO: what is the difference between training, test and test data sets;
- SO: neural network creation guide;
- SO: FAQ on neural networks;
Direct propagation neural networks
- SO: the role of offsets in neural networks;
- SO: choose the number of hidden layers and nodes;
- SO: choose the number of hidden layers and nodes;
- SO: choose the number of hidden layers and nodes;
- SO: simple neural network implementation;
Natural language processing
Support Vector Machine
- SE: the most popular questions about the support vector machine;
- SE: what is the support vectors method;
- SE: principles of the support vector machine;
- SE: comparison of support vector and classification trees;
- SE: support vector and logistic regression;
- SE: comparison of support vector and logistic regression;
- SE: when to use the support vector method, and when to use logistic regression;
- SE: what is the difference between support vector and logistic regression;
- SE: estimation of the importance of variables in the support vector method;
- SE: why do you need scatter scaling;
- SO: in which case the support vector machine is better than neural networks;
- SO: comparison of the support vector and neural networks;
Decision trees
- SE: weak side of decision trees;
- SE: how decision-based learning algorithms handle missing values;
- SE: decision trees are almost always binary, is this true?
- SE: what is the deviation;
- SO: comparison of algorithms for implementing decision trees by complexity or performance;
- SO: “pruning” tree branches in R;
- SO: how to extract the tree structure from the ctree function;
- SO: what is the entropy and the amount of information received when building decision trees;
Random forest
- SE: assessing the importance of variables in random forests;
- SE: comparing the R-squared value of two random forest models;
- SE: why a random forest does not handle missing values in predictors;
- SE: extract data from random forest algorithm;
- SE: questions on the implementation of a random forest on R;
- SO: implementation of a random forest on R;
- SO: questions on the implementation of a random forest on R;
Tree Boosting Algorithms
Compositional training
- SE: compositional models with caret;
- SE: bagging, boosting, stacking;
- SE: materials on the implementation of methods of compositional education;
Dimension Vapnik - Chervonenkisa
- SE: methods of compositional training increase the dimension of Vapnik - Chervonenkis;
[ Part 2 ]
')
Source: https://habr.com/ru/post/277511/
All Articles