User tdc

38 votes

Accepted

Bayesian lasso vs ordinary lasso

answered Jul 24, 2014 at 10:45

36 votes

Accepted

PCA, LDA, CCA, and PLS

answered Feb 20, 2012 at 8:31

30 votes

Is whitening always good?

answered Feb 16, 2012 at 11:11

26 votes

Why do naive Bayesian classifiers perform so well?

answered Feb 9, 2012 at 14:44

25 votes

Accepted

Is AdaBoost less or more prone to overfitting?

answered Jan 8, 2012 at 19:57

20 votes

How to decide on the correct number of clusters?

answered Feb 9, 2012 at 14:55

13 votes

Accepted

How would you design a machine learning system to play Angry Birds?

answered Feb 13, 2012 at 15:19

12 votes

Are there algorithms for computing "running" linear or logistic regression parameters?

answered Feb 9, 2012 at 9:45

11 votes

Why do we assume that the error is normally distributed?

answered Feb 10, 2012 at 9:44

10 votes

How can I help ensure testing data does not leak into training data?

answered Feb 15, 2012 at 17:14

10 votes

Accepted

Probabilistic graphical models textbook

answered Jan 11, 2012 at 14:46

10 votes

What are the measure for accuracy of multilabel data?

answered Dec 8, 2011 at 11:25

9 votes

Accepted

When over/under-sampling unbalanced classes, does maximizing accuracy differ from minimizing misclassification costs?

answered Dec 8, 2011 at 10:58

9 votes

Accepted

Does a sparse training set adversely affect an SVM?

answered Feb 10, 2012 at 9:05

8 votes

Accepted

Are MFCCs the optimal method of representing music to a retrieval system?

answered Feb 13, 2012 at 10:53

8 votes

Accepted

What are alternatives to VC-dimension for measuring the complexity of neural networks?

answered Feb 9, 2012 at 14:21

8 votes

What are the "hot algorithms" for machine learning?

answered Nov 14, 2011 at 10:49

7 votes

Accepted

Incremental IDF (Inverse Document Frequency)

answered Nov 25, 2011 at 10:37

7 votes

Accepted

Is Kernel Regression similar to Gaussian Process Regression?

answered Oct 28, 2016 at 22:52

6 votes

What is the proper naming scheme for dataset parts?

answered Feb 9, 2012 at 10:16

6 votes

Fast method for finding best metaparameters of SVM (that is faster than grid search)

answered Feb 8, 2012 at 9:41

6 votes

Why does Natural Language Processing not fall under Machine Learning domain?

answered Feb 20, 2012 at 8:43

6 votes

Why does ridge regression classifier work quite well for text classification?

answered Dec 7, 2011 at 14:13

6 votes

Accepted

Classification with tall fat data

answered Feb 8, 2012 at 9:55

5 votes

Accepted

Train a SVM-based classifier while taking into account the weight information

answered Jan 25, 2012 at 19:09

5 votes

Predicting multiple targets or classes?

answered Jan 30, 2012 at 14:32

5 votes

Functional principal component analysis (FPCA): what is it all about?

answered Feb 17, 2012 at 11:03

5 votes

Why use a train/test split with linear regression

answered Oct 20, 2017 at 21:07

4 votes

Accepted

Which classifiers work well with unbalanced data?

answered Jul 24, 2014 at 10:21

4 votes

What is a good method for designing an HTM network in terms of depth and breadth?

answered Feb 8, 2012 at 10:20

Stack Exchange Network

61 Answers

Bayesian lasso vs ordinary lasso

PCA, LDA, CCA, and PLS

Is whitening always good?

Why do naive Bayesian classifiers perform so well?

Is AdaBoost less or more prone to overfitting?

How to decide on the correct number of clusters?

How would you design a machine learning system to play Angry Birds?

Are there algorithms for computing "running" linear or logistic regression parameters?

Why do we assume that the error is normally distributed?

How can I help ensure testing data does not leak into training data?

Probabilistic graphical models textbook

What are the measure for accuracy of multilabel data?

When over/under-sampling unbalanced classes, does maximizing accuracy differ from minimizing misclassification costs?

Does a sparse training set adversely affect an SVM?

Are MFCCs the optimal method of representing music to a retrieval system?

What are alternatives to VC-dimension for measuring the complexity of neural networks?

What are the "hot algorithms" for machine learning?

Incremental IDF (Inverse Document Frequency)

Is Kernel Regression similar to Gaussian Process Regression?

What is the proper naming scheme for dataset parts?

Fast method for finding best metaparameters of SVM (that is faster than grid search)

Why does Natural Language Processing not fall under Machine Learning domain?

Why does ridge regression classifier work quite well for text classification?

Classification with tall fat data

Train a SVM-based classifier while taking into account the weight information

Predicting multiple targets or classes?

Functional principal component analysis (FPCA): what is it all about?

Why use a train/test split with linear regression

Which classifiers work well with unbalanced data?

What is a good method for designing an HTM network in terms of depth and breadth?