What courses do you want to see the deeplearning.ai team build next?
Reinforcement learning, unsupervised learning in all its flavours and applications by TensorFlow shoulb be very appreciated.
I think the style used in DL courses of Andrew Ng is a perfect mix of theory and practice and should be taken as an example.
In my opinion in every course more time should be spent on Keras and TensorFlow exercises because they've not a friendly API interface and are very tough to master!
I see a lot of people are interested in RL but I think there are many other important things that needs to be covered first. RL still has very very limited applications in the real world. It is good to explore but the applications are very limited at this point in time. Here are few things that I think are more important to cover:
- Advanced DNN architectures: MobileNets(v1 and v2), ResNeXt, WideResNets, etc.
- Advanced object detection: SSD, YOLOv3, RetinaNet. How it is implemented in SDCs?
- Segmentation: UNets, DeepLab(v1, v2 and v3), DenseNet. Things in medical imaging and self driving cars
- Optimization algorithms: Learning rate finder, Cyclic learning rates, One cycle policy, super-convergence, AdamW
Recommendation engines using Deep learning
NLP and Speech: BERT, UMLfit, etc.