Machine Learning Control: Genetic Programming Control

[@fackarov9412]

one of the best courses never seen, thx

[@yiyangwu1314]

Thx~ prof Appreciate it for your sharing We really need these intelligent lectures

[@gxprestes]

Hello Professor Steven, Thank you very much for sharing this approach. I am very excited about it. However, I am unable to use the GP as described in your book because the provided repository contains errors, and the repository link in the video description pertains to your other book. I would be extremely grateful if you could share the example comparing GP with LQR as demonstrated in your book. Wishing you all the best!

[@sophiakim107]

Hi Prof B. with genetic programming control, my understanding is that you can directly find control laws to manipulate your system. Is it correct that you don't necessarily need to identify your system dynamics in order to apply this technique?

[@Eigensteve]

That is absolutely right, and that is one of the main advantages for complex systems like turbulence.

[@dsfsddude9336]

Thanks Prof. I know it is a old video but maybe you see this: What can you pull from the genetic programming control ... interpretation wise? Like you can probably see what parameters are included and the associated parameter values for the control, but can this inform anything about how the system dynamics itself look? I suppose my main question is whether we could infer aspects of system dynamics purely from the knowledge we gather on how effective control looks...

[@dsfsddude9336]

Well I guess in a way this would be related to OKID/ERA or SINDYc and so on... still would love your input

[@RooftopDuvet]

This reminds me of the relationship of compressed sensing to measurement... but it's the dual - it's compressed actuation.