Optimizing a step function of two bounded variables

What is the best matlab function for optimizing a step function of two bounded variables?

Edit: I don't know if it makes any difference, but the step function is bounded as well (in [0,N] for N known)

[Athul Prakash]

What do you mean by step function? Does it step only once at x=0 or does it step multiple times? Please specify what kind of function you are asking about. I am assuming you are handling this as a matlab function (either a file or local function). Could you clarify this as well?

[Antonios Nasioulas]

Hello and sorry for the late reply.

It steps multiple times. I am doing some simulations of a reinforcement learning task using argmax and I want to do parameter revovery. Because I use argmax, instead ot miniminizing lik = - Σ log(P(choice)) I want to minimize lik = -Σ (P(choice)) [the number of choices explained]. This gives a step function.

I didn't get what is your second question. If you could please clarify it.