Systems for of automatically classifying visual content in images has improved substantially in the last 2-3 years due to the introduction of methods of “deep learning”. This is based on training a computational network using large sets of labeled image data, i.e. where the content of the image has been annotated manually. I will speak about this and possible future developments in vision and other fields.
– Why is there a breakthrough in AI right now?
– What is the main idea with deep learning and why does it work?
– What will happen in the future?