Solve a Dynamic Programming optimization problem using Neural Networks

Good afternoon,

I have a dynamic programming problem (closed-loop, with three state variables), where each period's decision will directly impact other period's, and so they should be taken into account, with a series of constraints. Essentially, the goal is to define a rule within each period which can decide whether or not to perform action A, based on previous and future states, and a certain associated cost.

Is it possible to do the same problem using Neural Networks?

[Greg Heath]

Need more info

Greg