Export the reinforcement learning result - to see the weights of critic network and actor network

Matlab Help document gives an example about the DDPG applying to the double integrator system.

However, after training the agent, it seems that the weights cannot export.

How can I see the weights?

Moreover, can the DDPG agent apply to the time-varying control system ?

Answer 1

[Emmanouil Tzorakoleftherakis]

Hello,

You can see the values of the neural network weights using this function.

Yes, you can apply DDPG and RL in general to time-varying problems, but you need to be careful how you set up your training. For example you may want to consider adding noise to the time-varying reference to avoid overfitting to a specific scenario.