Indian Institute of Technology Madras
BS Degree Programme · Examination & Study Paper
Mathematical Foundations of Generative AI
VAEs & reparameterisation
Official Paper
Examination Instructions & Guidelines
- Solve on notebook/paper first before checking solutions.
- MCQ: Exactly 1 option is correct. Negative marking applies.
- NAT: Numerical response. No negative marking.
- Scientific calculator allowed for all calculations.
- Official answer key is sealed at the end of this sheet.
End Term · 31 Aug 2025 (FN)End Term
- Q1MCQ4 Marks·Choose one option
- A
- B
- C
- D
- A
- Q2MCQ4 Marks·Choose one option
- A
- B
- C
- D
- A
- Q3MCQ5 Marks·Choose one option
In the Sampling pipeline in VAE, Unconditional generation uses
- A
- B
- C
- D
- A
End Term · 31 Aug 2025 (AN)End Term
- Q4MCQ4 Marks·Choose one option
- A(1.205,−3.210)
- B(1.205,−1.895)
- C(0.500,−3.210)
- D(0.705,−2.000)
- Q5MCQ4 Marks·Choose one option
Given d_z=20, if the first decoder layer is a fully connected layer that outputs 3 x 64 x 64 units and includes a bias term, how many trainable parameters does this layer have?
- A245,760
- B258,048
- C122,880
- D262,144
Quiz 2 · 3 Aug 2025 (AN)Quiz 2
- Q6NAT2 Marks·Write the number
Your answer: - Q7NAT3 Marks·Write the number
Your answer: - Q8NAT3 Marks·Write the number
Your answer: - Q9NAT4 Marks·Write the number
Your answer: - Q10MCQ3 Marks·Choose one option
- A
- B
- C
- D
- A
- Q11MCQ3 Marks·Choose one option
- A
- B
- C
- D
- A