Optimizing a step function of two bounded variables

2 次查看（过去 30 天）

显示更早的评论

Antonios Nasioulas 2020-3-31

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/514333-optimizing-a-step-function-of-two-bounded-variables

编辑： Antonios Nasioulas 2020-4-8

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)

2 个评论
显示无隐藏无

Athul Prakash 2020-4-3

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 2020-4-8

编辑：Antonios Nasioulas 2020-4-8

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.

请先登录，再进行评论。

请先登录，再回答此问题。