Deep learning is machine learning with neural networks made of many layers. It powers language models, image recognition and speech recognition.
Example
A network learns to recognise cats from millions of photos without anyone writing rules about ears or whiskers.
For pros
Each layer learns more abstract features from the output of the previous one. The breakthrough from around 2012 came from large datasets, GPUs and better training methods. Deep learning needs a lot of compute and is hard to explain.
