Explain overfitting, underfitting, and regularization

You are asked ML fundamentals questions. 1. What are **overfitting** and **underfitting**? Describe how they show up in **training vs. validation/test** performance. 2. Explain the **bias–variance tradeoff** and how it relates to over/underfitting. 3. What is **regularization**? Compare **L1** and **L2** regularization: - effect on weights - effect on sparsity / feature selection - when you might prefer one over the other 4. What is **dropout**? Why can it reduce overfitting, and what are common pitfalls when using it? 5. List a few additional techniques to reduce overfitting (besides L1/L2/dropout) and when you would use them.