3 ms·Re-enforcement learning would be fun on this one.by dzink 6y agoRe-enforcement learning would be fun on this one.