Indian Institute of Technology Madras
BS Degree Programme · Examination & Study Paper
Mathematical Foundations of Generative AI
GAN objective & training
Official Paper
Examination Instructions & Guidelines
- Solve on notebook/paper first before checking solutions.
- MCQ: Exactly 1 option is correct. Negative marking applies.
- MSQ: One or more options correct. Select all to score.
- NAT: Numerical response. No negative marking.
- Scientific calculator allowed for all calculations.
- Official answer key is sealed at the end of this sheet.
Quiz 1 · 26 Oct 2025 (AN)Quiz 1
- Q1MCQ2 Marks·Choose one option

- A−log 2
- B0
- C−log(0.3)
- Dlog 2
- Q2MCQ2 Marks·Choose one option

- A80,402
- B99,402
- C242,503
- D159,3
- Q3MSQ3 Marks·Choose all correct options
Which of the following are valid reasons why GAN training based on JSD minimization is unstable?
- AJSD saturates when the supports of p_(data) and p_g do not overlap.
- BGradients for the generator can vanish when the discriminator becomes too strong.
- CJSD is not continuous w.r.t. the parameters of the generator distribution.
- DJSD always provides unbiased gradients in high dimensions.
- Q4NAT2 Marks·Write the number
Your answer: - Q5NAT3 Marks·Write the number
Your answer:
End Term · 31 Aug 2025 (FN)End Term
- Q6MCQ3 Marks·Choose one option

- A

- B

- C

- D

- A
- Q7MCQ4 Marks·Choose one option
Starting from 4×4 spatial size, apply four ConvTranspose2d blocks each with kernel 4, stride 2, padding 1. Final spatial size is:
- A16 × 16
- B32 × 32
- C64 × 64
- D128 × 128
End Term · 31 Aug 2025 (AN)End Term
- Q8MCQ3 Marks·Choose one option

- AZero a.e.
- BNonzero and large
- CUndefined
- DEqual to the score of p_g
- Q9MCQ4 Marks·Choose one option

- A31
- B32
- C33
- D34