Follow up video on the dog learning to algorithmically extract a ball from a machine to pay for treats, making random solutions not work as easily anymore and showing more learning progress