Thanks Udit.

DL is still weak at interpretability, though there are some initial advances. If explainability is crucial, it’s better to go for simpler ML models.

Its best to explore simpler techniques/models first. I had addressed this in my most recent story (pasting again below):

You may wonder when deep learning must be used vis-a-vis other techniques. Always start with simple analysis, then probe deeper with statistics, and apply machine learning only when relevant. When all these fall short, and the ground is ripe for some alternate, expert toolsets, dial in deep learning.