I have two equal length sequence, with a set of computational tasks sequence [1, 2, 3, 4, 5 6], a set of server order [1, 2, 3, 4, 5 6], each task matching a server, fixed the server after the order, on the ordered by computing tasks such as [5 4 3 2 1 6] such a sequence into a computational algorithm for energy consumption, want to make the calculated minimum energy consumption, if use A3C this action how to define? Or with any other machine learning algorithms is better to solve this problem?