Linear algebra. Bayesian statistics. MUST know these inside out, upside down.
Vector calculus. Convex optimization.
A boatload of machine learning literature. The ideas coalescing into deep learning are based more than a decade of research.
If you know nothing about math... I can't imagine getting to the point of understanding deep learning (which is a fairly rapidly evolving area) without at least 2-3 years of very hard work.
This class is a reasonable attempt to give a quick intro to one major source for DBNs https://www.coursera.org/course/neuralnets understanding this course is a good benchmark.
Comments
Math. math. math.
Linear algebra. Bayesian statistics. MUST know these inside out, upside down.
Vector calculus. Convex optimization.
A boatload of machine learning literature. The ideas coalescing into deep learning are based more than a decade of research.
If you know nothing about math... I can't imagine getting to the point of understanding deep learning (which is a fairly rapidly evolving area) without at least 2-3 years of very hard work.
This class is a reasonable attempt to give a quick intro to one major source for DBNs https://www.coursera.org/course/neuralnets understanding this course is a good benchmark.