faq: What is the difference between a *cost function* and a *loss function* in machine learning?

Author: rasbt

README.md

@@ -184,6 +184,10 @@ I have set up a separate library, [`mlxtend`](http://rasbt.github.io/mlxtend/),
 
 - [What are the advantages of semi-supervised learning over supervised and unsupervised learning?](./faq/semi-vs-supervised.md)
 
+##### Ensemble Methods
+
+- [Is Combining Classifiers with Stacking Better than Selecting the Best One?](./logistic-boosting.md)
+
 ##### Preprocessing
 
 - [Why do we need to re-use training parameters to transform test data?](./faq/scale-training-test.md)

faq/README.md

@@ -107,6 +107,10 @@ Sebastian
 
 - [What are the advantages of semi-supervised learning over supervised and unsupervised learning?](./semi-vs-supervised.md)
 
+##### Ensemble Methods
+
+- [Is Combining Classifiers with Stacking Better than Selecting the Best One?](./logistic-boosting.md)
+
 ##### Preprocessing
 
 - [Why do we need to re-use training parameters to transform test data?](./scale-training-test.md)

faq/logistic-boosting.md

@@ -0,0 +1,10 @@
+# Do bagging and boosting can be used with logistic regression?
+
+I am not sure if bagging would make much sense for logistic regression -- in bagging, we reduce the variance of the deep decision tree models that overfit the training data, which wouldn't really apply to logistic regression.
+ 
+Boosting could work though, however, I think that "stacking" would be a better approach here. Stacking would be more "powerful" since we don't use a pre-specified equation to adjust the weight, rather, we train a meta-classifier to learn the optimal weights to combine the models.
+  
+  
+Here's one of the many interesting, related papers, I recommend you to check out :)
+
+- "Is Combining Classifiers with Stacking Better than Selecting the Best One?"