The Problem<p>Let's say that we have a machine that spits out a number (between 1 to 15) alongside a color (either green or blue) every 10 seconds.<p>OK so every 10 seconds we get a number (between 0 to 15) and a color (green/blue), cool, BUT we do not have any clue how these choices are made by the machine, we can just observe the output and record it at best.<p>The Question<p>can we use ML to feed in a model with a lot of these numbers and/or colors as they are put out by the machine and after a while let the model predict the next number and/or color which is going to be generated by the machine?