~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This is a Machine Learning Prototype (MLP) made by me. Please read the description in full. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This algorithm can: 1. Get to the Goal reliably, no matter where it is 2. Learn how to get to the Goal very quickly 3. Receive both rewards and Punishments This algorithm cannot: 1. Avoid difficult obstacles 2. Determine 100% accurately when and when not to give rewards (if its touching the goal, it will always give rewards, which can be a problem) 3. Have the goal change mid learning and it still work 4. Do anything besides move in a direction ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This is a PROTOTYPE and PROOF OF CONCEPT, I have always found machine learning to be fascinating and wanted to try my hand at it on scratch. This is version one, with more to come. Many, many more. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Learn how the algorithm works by clicking See Inside! You can also move the goal or even try and make your own conditions for rewards! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
