Day 19 of 133

Decision trees + DSA Binary Search

Splits (Gini/entropy), pruning, why trees are high-variance.

DSA · NeetCode Binary Search

Find Minimum IN Rotated Sorted ArrayDSA · Binary Search
Interview questions to prep
1. Why does mid-vs-right comparison work but mid-vs-left can mislead with duplicates?
2. How does the algorithm change if duplicates are allowed?
Search IN Rotated Sorted ArrayDSA · Binary Search
Interview questions to prep
1. How do you decide which half is sorted on each step? Walk through with a small example.
2. What changes if duplicates are allowed (LC 81)?

How a decision tree splits: Gini, entropy, information gainTraditional MLStatQuest
Interview questions to prep
1. How does a decision tree decide where to split? Compare Gini vs entropy.
2. Why are decision trees high-variance, and how does pruning help?
Trees vs linear models: when each winsTraditional MLRead
Interview questions to prep
1. When does a tree-based model crush a linear model and vice versa?
2. Why don't decision trees need feature scaling?
Feature importance: gain, permutation, SHAPTraditional MLMolnar
Interview questions to prep
1. Compare gain-based feature importance, permutation importance, and SHAP — pros and cons.
2. Why can gain-based importance be misleading when features are correlated?

References & further reading