*simple explanation* An AI has to learn to get as close to the green dot as possible in 10 steps. It learns to do this gradually, so you have to wait for 10-15 generations to see some results This project uses a modified version of Scratchnet 1.4 (https://scratch.mit.edu/projects/238232733/) Scratchnet is a fully customizable neural network made for others to use in their machine learning projects. The current example has a 4-5-5-2 neuron layout. It takes its position and the dot position as inputs and has to output a direction and an distance. After 10 steps its fitness is calculated. Scratchnet is relatively easy to use yourself and you can always ask me for support if you need it.
Current version: 1.4 All made by me using my own knowledge during a 14 hour car trip. This project is completely open source (all Scratch projects are in theory) but credits are always appreciated. See inside for more info.