Today, machine learning techniques can be used to analyze an image and find out which object(s) are visible on the image. I wanted to find out how this works and designed a Convolutional Neural Network, which divides images into 2 classes, depending on whether a dog or a cat is shown. This blog post describes the development and discusses the results.
Animals don’t ask themselves: What if? They also use tools, but no bird will ever use a stick to try to get a stone out of a hole just to be prepared if one day in the future there is a nut instead of the stone.
Machine Learning is more than neural networks. This blog post provides a quick overview and links to the most important concepts.