Hi All,
I was wondering if anyone in the community could offer some advice on the latest developments in model explainability, interpretability, and simplification?
I’m currently working on a project where we are training Deep Neural Networks to predict a health condition, so explainability and interpretability are critically important for adoption.
I discovered the PiML toolkit from another post in a DeepLearning.ai forum, and it’s been brilliant as a tool for exploring and interpreting models. It also led me to read this paper which provides a method for simplifying Deep Neural Networks which I think is likely to make the resulting models more acceptable for use in the healthcare field.
However I am not an ML/AI researcher, and I was wondering if anyone could point other significant libraries, papers, or projects that I should take a look at?
Also thank you to the educators, moderators and other staff involved in delivery of the DeepLearning.ai courses. I recently completed the Deep Learning Specialisation and it was awesome!