Machine Learning: Good book for learning Probability and statistics? I am trying to learn machine learning and looking for a good book to understand probability and statistics from machine learning point of view and for the sake of understanding probability. Though I have studied probability in the past, I am still having a hard time to solve homework questions from Stats 110 course by Blitzstein. I think I am missing many concepts in probability theory as I didn't pay attention in my Probability classes.
So what I need is a good book which can kind of introduce me and at the same time refresh some concept left in my brain and provides good intuitive explanation of questions and their answers and provide very good but few important exercises to understand all the concepts. I checked many questions like this and I am still in a dilemma which book to consider: 
1 - An Introduction to Probability Theory and Its Applications - William Feller
2 - A First Course in Probability - Sheldon Ross
Which one would you recommend? Or if you have other good book in mind please do let me know. 
 A: Of the two books you mention, I have read both, and for actually learning probability, A First Course in Probability by Sheldon Ross is definitely better for a first book. In particular, I found the problems to be the best of any comparable introductory book on probability theory.
Feller has many good insights and his writing style is enjoyable at times, but I don't feel that the problems aren't well-suited for someone new to the field. Also it is arguably too wordy for someone who needs to learn, practice, and become acquainted with the theory for the first time. Personally I think it is better to read "for culture" after having already learned probability for the first time. 

If you are interested in machine learning in particular, I recommend you consider the following two books:
1. Bishop's Pattern Recognition and Machine Learning, freely available here,
2. Murphy's Machine Learning: A Probabilistic Perspective.
Both books begin with thorough introductions to the probability theory and statistics relevant specifically to machine learning, before addressing machine learning itself. They both seem well-suited to what you are looking for, especially the latter, if you want to understand probability and statistics from the machine learning point of view.
A: The three-volume series by Hoel, Port, and Stone is good. Does not expect much background, and there is a clear distinction between probability, statistics, and stochastic processes ( a separate volume is devoted to each topic). Moreover, each volume is rather short with good problems sets.
Other than that heard good things about Grimmett and Stirzaker and its exercise book.
A: Python for Probability, Statistics, and Machine Learning is focused on an intuitive grasp of these topics. Most of the book is available on github as IPython/Jupyter notebooks 
