Pattern recognition and machine learning by Bishop is one of the canonical text books. It helps to have a linear algebra background, it includes a refresher though
Bishop is good but reads a little too much like a literature review sometimes. That may or may not be a problem depending on what you are looking for.
Thanks everyone for the suggestions...will check these books out
This one is interesting to see the "statistical" other side of the industry vs machine-learning people. For example I don't think gradient descent is used once in that book.
This book as a prerequisite to anyone who wants to get into machine learning imo.