matthieukomorowski / AI_Clinician

Reinforcement learning for medical decisions
99 stars 46 forks source link

Question about figure 2.b #15

Open Uwonsang opened 4 years ago

Uwonsang commented 4 years ago

When I implement your Matlab code, I troubled with fig.2.b.

In your paper, AI Estimated policy value of fig2.b is generally high.

but AI estimated policy value of My fig.2.b looks like Random estimated policy value of original fig.2.b.

Original fig.2.b

figure1

My fig.2.b figure2

My figure doesn`t look like the original one.

please check the fig2.b code,

if their problem, Please Reply to me.

huangtao36 commented 4 years ago

I got the same result as you