Explain overfitting, imbalance, undersampling, and attention heads

How do you diagnose and mitigate overfitting in machine learning models? Discuss techniques such as regularization (L1/L 2), data augmentation, early stopping, dropout, architecture changes, cross-validation, and appropriate validation strategies; explain trade-offs for each. How do you handle class imbalance? Compare undersampling, oversampling (including SMOTE/variants), class weighting, focal loss, and thresholding; specify when each is appropriate and how to evaluate with metrics like precision/recall, PR-AUC, ROC-AUC, and calibration. Detail common undersampling strategies—random undersampling, Tomek links, Edited Nearest Neighbors (ENN), NearMiss variants, and cluster-centroid methods—and discuss their impact on bias/variance and minority class recall. What is an attention head? Explain queries, keys, and values; how multi-head attention splits representation space; what multiple heads capture; and how head count affects capacity, compute, and inductive bias.