Prof. Meena's Exam Prep
Disclaimer: This project is not intended to cover all material that may appear on the exam; rather, it serves as a foundational starting point. If any content on this site is unclear, we encourage you to conduct further research or consult with your peers.
Calculus
The Prerequisite. Chain Rule & Derivatives needed for Backpropagation.
Linear Algebra
The Foundation. Matrix Representation & Operations that serve NNs.
LeetCode & DP
Problem Solving. Basic DP and logic patterns (easier than Algorithms course).
Reasoning For The Selected Topics
1. Linear Algebra: "Serving the NN"
The professor is teaching NNs, so he definitely wants the math that serves his subject. He won't ask random theory, he will most likely ask about Matrix Representation and Operations.
- Matrix Multiplication: The core operation.
- Representation: How data (inputs/weights) is stored in matrices.
2. Calculus: "The Tool for Backprop"
He said he won't ask about the deep details of the course itself (like implementing Autodiff engines) because he hasn't explained them yet. He most likely wants the Chain Rule because that's the prerequisite.
Every gradient flow is just multiplying local derivatives.
If L is Loss function in terms of c, and c = b², then ∂L/∂b = (∂L/∂c) · 2b
3. LeetCode: "Basic Logic"
Logically, the coding problems will be easier than a dedicated Algorithms course. He likely wants to test Basic DP and famous pattern problems.
- Famous Problems: Move Zeroes, Transpose Matrix.
- Basic DP: Simple optimization logic (like Stock Buy/Sell).
Small Reminder
"Dr. Meena wants the ingredients, not the cake (yet)."
Master the Chain Rule (for Calculus) and Matrix Dimensions (for Linear Algebra), and Inshallah you will be ready for whatever he throws at you.