How much math did you have to take to get your CS degree?
How much did you take as electives?
I took quite a few elective courses such as real analysis, abstract algebra and stochastic processes.
In that case you might be interested to know that backpropagation is a lot easier if you are good at differential equations, and implementing a neural net is much more efficient if you know linear algebra.
Really, pick up an AI book. It's all math.
You don't need differential equations for that. You need DIFFERENCE equations. Those are used in the Kalman filter which is what is used in estimation of states. Also for AI the only math you need is logic, set theory and probability&statistics.
Up to and a focus on modern algebraic geometry and arithmetic geometry.
Graph theory: cryptography; math logic; linear, nonlinear, and discrete optimization; category theory; everything dealing with geometry.
Calc I, II, III, Diff eq (I have no idea why so much focus on this, should be more on discrete)
Discrete I, II (graph theory, various combinatorics (generating functions, etc), random other shit thrown in like formal logic and shit)
Probability/Statistics (although I wouldn't count this really)
more graph theory
abstract algebra (I think having this required instead of one of those required calc courses might actually benefit more, although so would more graph theory)