3
$\begingroup$

For personal reasons, I need a recent reference book on ML which is not mainly focused on programming, such as for example

https://www.amazon.com/Hands-Machine-Learning-Scikit-Learn-TensorFlow/dp/1492032646/

but which is also not overly theoretical, such as

https://www.amazon.com/Understanding-Machine-Learning-Theory-Algorithms/dp/1107057132/

I discarded Bishop's PRML because it's 10 years old now. I was thinking to choose among:

https://www.amazon.com/Machine-Learning-Probabilistic-Perspective-Computation/dp/0262018020/

https://www.amazon.com/Elements-Statistical-Learning-Prediction-Statistics/dp/0387848576/

https://www.amazon.com/gp/product/0262028182/

I value clarity of exposition and notational consistency over completeness. Price is not an issue. I'm open to other suggestions, if you think there are better options than the three I listed.

PS this one looks very nice too

https://mml-book.github.io/

but it hasn't still been published. Also, even though I'm not looking for the most complete treatise, this does seem a little short.

$\endgroup$
3
  • $\begingroup$ It is really hard to say what you are looking for. A book you want to use as a reference for research? A book for self studying? Does it have to be a book or are lecture notes enough? Why does it have to be new? $\endgroup$ Commented Nov 2, 2019 at 20:10
  • $\begingroup$ 1. I was thinking of a reference book, since I'm a researcher. However, as I said I prefer a book which is clear (and uses a consistent notation) over some 1000 pages behemoth, which maybe lists every possible paper, but it's dispersive or hard to read. 2. I prefer books to lecture notes (especially if they're in PowerPoint) but particularly good lecture notes may be considered. 3. Why not? $\endgroup$
    – DeltaIV
    Commented Nov 3, 2019 at 16:08
  • $\begingroup$ I personally like the elementa of statistical learning that you already mentioned. But depending on what specific area of ML you are interested in affects things. Deep learning? Mostly time series related? Very mathematical? Some other specific method? Maybe graph related data? $\endgroup$ Commented Nov 4, 2019 at 17:21

2 Answers 2

2
$\begingroup$

Machine Learning Refined: Foundations, Algorithms, and Applications by Jeremy Watt, Reza Borhani, Aggelos Katsaggelos. This recent book strikes an excellent balance between mathematics, figures and code.

$\endgroup$
2
$\begingroup$

I would recommand the new (2022) version of the Probabilistic ML book from Kevin P. Murphy: https://probml.github.io/pml-book/book1.html.

I am a huge fan of the approach from the first version: it has a very clear introduction, taking time on notations and construction of the framework. Then, all examples are clearly explained with different levels of complexity, making it easy to navigate.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.