I hooked a neural network up to my Roomba. I wanted it to learn to navigate without bumping into things, so I set up a reward scheme to encourage speed and discourage hitting the bumper sensors.— Custard Smingleigh (@Smingleigh) November 8, 2018
It learnt to drive backwards, because there are no bumpers on the back. https://t.co/8PdR3p6ePZ
They *love* vacuuming up cornflakes— Custard Smingleigh (@Smingleigh) November 8, 2018
(The neural network has a reward function which tells the Roomba whether or not it's doing a good job, and when the Roomba is not doing a good job the neural network learns to do less of the thing it's currently doing)