What should I learn if I want to make a decision based on multiple variables? Followings are the example of a problem.

I have a farm. My variables are weather, humidity of air, humidity of soil, size of my fruits, etc.. Every hour, I will capture every data and store it in my database. My possible decisions are "Harvest", "Plant", "Wait". Sorry if this question is too generalized or too broad. I do not know what I want to learn.

  • $\begingroup$ It sounds like a control problem with continuous state space. You might want to look into Reinforcement Learning. $\endgroup$ – Shahaf Finder Feb 13 '20 at 14:38
  • $\begingroup$ @ShahafFinder thank you for the recommendation. I have read some text regarding on my problem. Turns out there are no ways (yet?) to solve my issue since it is dealing with uncertainties and white noises. $\endgroup$ – dude Feb 13 '20 at 15:07
  • $\begingroup$ If so, it will be interesting to see how well can you approximate it. If it interests you, don't give up :) Maybe try solving an easier problem first (say, a simulated problem with no noise). $\endgroup$ – Shahaf Finder Feb 13 '20 at 15:14
  • $\begingroup$ @ShahafFinder but how can one make a decision if they do not even know that they do not have enough variable or not (hence the white noise).. to be honest, I think my research is stalling because I don't think I have enough variables to include in my statistics, but I just dont know what it is or how to find it. I feel so defeated. $\endgroup$ – dude Feb 13 '20 at 15:24
  • $\begingroup$ Maybe you can ask in Cross Validated, they might have a better idea of these kind of stuff. stats.stackexchange.com $\endgroup$ – Shahaf Finder Feb 13 '20 at 20:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.