0

Is there a book analogous to Introduction to Algorithms (CLRS) for Machine Learning, that covers all the models and explanations comprehensively?

sloth
  • 141
  • 4
  • Just to clarify, when you say "all the models" do you just mean all types of neural nets? – Hugh Nov 05 '16 at 03:38
  • No, everything from rudimentary logistic regression to deep learning. – sloth Nov 05 '16 at 12:49
  • 3
    I don't think such a book exists. I certainly don't know it. Deep learning is an area of active ongoing research, so there can't be an encyclopedic reference on that. If you search amazon "deep learning' in books, the 3 top (actual) deep learning books are yet to be released. For topics other than deep learning 2 standard and (imho) very good references are Elements of Statistical Learning by Hastie, Tibsharani, and Friedman and Machine Learning a Probabilistic Perspective by Murphy. – meh Nov 05 '16 at 13:00
  • @aginensky +1 for Hastie – JAD Nov 05 '16 at 13:01

1 Answers1

0

My recommendation is Applied Predictive Modeling by Max Kuhn. This book covers all the major regression and classification models, as well as some lesser known models, such as Cubist and C5.0. While the text does not cover deep learning, it does cover caret, which is Kuhn's R package. The caret package contains several deep learning models (e.g. stacked autoencoders) as well as many other classification and regression models.

Dirigo
  • 376
  • 3
  • 8
  • This is a great book (I have even bought a copy of it) but I do not think it is good for what the OP wants. It is far too applied to be assumed that it "*covers all the models and explanations comprehensively*". – usεr11852 Nov 05 '16 at 16:36