MATLAB Answers

reinforcement learning using my own function

1 view (last 30 days)
Adam Horvath
Adam Horvath on 3 Feb 2020
I have just started dabbling in machine learning. I would like to use reinforcement learning to do the following: I have a "core" function that takes 5 input numbers and returns 1 output number. I would like to use a reinforcement learning algorithm to input the 5 numbers into my function and try to maximize the 1 output number. Does my function qualify as "environment"? I am feeling a bit lost in this topic. Could someone point me in the right direction as to which toolbox to use and maybe the functions I should look at in the relevant toolbox?
Thanks a lot!

Answers (1)

Emmanouil Tzorakoleftherakis
It looks like your "core" function qualifies as the actual policy (or value function for that matter). The environment would be the thing that decides how these 5 inputs to the policy are generated. If you share more info on the application it may be easier to provide more suggestions.

Community Treasure Hunt

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

Start Hunting!