Your question is Overfitting and Underfitting in Deep Learning. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You are training a deep learning model for a supervised learning task and comparing several architectures and training settings. You want to understand why some models fail to generalize while others never learn enough signal from the data.
What are the primary causes of overfitting and underfitting, and how do you diagnose and mitigate them in deep learning models?