Reinforcement Learning Agents generating zero episode

1 次查看（过去 30 天）
Hamid Fazeli
Hamid Fazeli 2022-10-1
评论： Hamid Fazeli 2022-10-19
Hello Matlab community,
I faced an issue for training a multiple agent problem in Matlab/Simulink. I tried to solve a very simple problem; however, training stops at episode 1.
Suppose that we have three discrete variables A=[1 2 3], B=[1 2 3], C=[1 2 3].
Reward function = A*B*C;
Observation= A+B+C;
I tried different parameters but it didn’t work. I attached the sample file for the reference. Very appreciate it if you can suggest potential solution to solve convergence issue.
Thanks for your time 😊

请先登录，再回答此问题。

采纳的回答

Ari Biswas
Ari Biswas 2022-10-4
There is an issue with the way you specified the reset function. Your function resetRobots should return a Simulink.SimulationInput object, which is also an input argument to the function. So for e.g. the correct function signature should be:
function in = resetRobots(in, var1, var2, var3)
% write reset code
end
See this example to see how reset functions can be defined.
https://www.mathworks.com/help/reinforcement-learning/ug/water-tank-reinforcement-learning-environment-model.html#WaterTankEnvironmentModelExample-5

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Reinforcement Learning 的更多信息

标签

产品


版本

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by