I would like to try a Reinforcement Learning approach for a multi-label or binary classification of a CSV file. I know that Supervised Learning is probably easier and I have already tried a couple of approaches. I works pretty good, but this RL approach could be an interesting thing for a small chapter and the outlook of my Master thesis. So i just want to try it even if the results might be much worse.
I found a couple of papers about image classification using RL and it seems to be possible. So i don't see a reason why it shouldn't work for a numerical Dataset. I could use the difference of the prediction/action of the RL agent and the actual label to train the model.
My problem is that i am not that good in coding. I have never done Reinforcement Learning and don't realy know where to start. Has anyone tried to use RL for classification or knows a good publication with a rather simple code attatched that i could take a look at?