MauryaHub PYQ Practice

da5002_2025T2_Q1_NA.pdf

Mathematical Foundations of Generative AI · Quiz 1 · May 2025

← Course papers · Start practice / exam

Questions and published explanations below are available without starting a test. Some questions may not have a published solution yet.

Question 189 NAT · 1.5 marks

You have a visual codebook with 1500 visual words. From a test image, you extract 1200 SIFT descriptors, each of 128 dimensions. After assigning each descriptor to its nearest visual word, you compute the Bag-of-Words histogram by counting descriptor assignments. Based on the above data, answer the given subquestions.
What is the dimensionality of the Bag-of-Words (BoW) feature vector for this image?________________

    A published solution is not available for this question yet.

    Question 190 NAT · 1.5 marks

    You have a visual codebook with 1500 visual words. From a test image, you extract 1200 SIFT descriptors, each of 128 dimensions. After assigning each descriptor to its nearest visual word, you compute the Bag-of-Words histogram by counting descriptor assignments. Based on the above data, answer the given subquestions.
    If instead, you concatenate all the 1200 descriptors into a single vector (without using BoW), what will be the dimensionality of that vector?__________________

      A published solution is not available for this question yet.

      Question 192 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_3_0]]
      Source diagram or notation
      1. [[IMAGE:ce01f99902ff74dc_3_1]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_3_2]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_3_3]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_3_4]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 193 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_3_5]]
      Source diagram or notation
      1. [[IMAGE:ce01f99902ff74dc_3_6]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_3_7]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_3_8]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_3_9]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 194 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_4_10]] Where K is an integer. What does this loop imply?
      Source diagram or notation
      1. Generator is updated multiple times per discriminator update.
      2. Discriminator is updated multiple times per generator update.
      3. Both networks are updated simultaneously.
      4. The model uses gradient accumulation.

      A published solution is not available for this question yet.

      Question 195 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_4_11]]
      Source diagram or notation
      1. The generator is overfitting to the noise.
      2. The discriminator has become too strong.
      3. The learning rate of the generator is too high.
      4. The noise vector dimension is too large.

      A published solution is not available for this question yet.

      Question 196 MCQ · 2.0 marks

      In Bi-GAN, which of the following expressions represents the correct joint objective to be minimized by the encoder E and generator G, and maximized by the discriminator D?
      1. [[IMAGE:ce01f99902ff74dc_5_12]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_5_13]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_5_14]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_5_15]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 197 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_5_16]]
      Source diagram or notation
      1. [[IMAGE:ce01f99902ff74dc_5_17]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_5_18]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_5_19]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_5_20]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 198 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_6_21]]
      Source diagram or notation
      1. It is finite and continuous even when pr and pg have disjoint supports.
      2. It is always zero for disjoint supports.
      3. It upper bounds the KL divergence.
      4. It requires the same dimensionality of supports.

      A published solution is not available for this question yet.

      Question 199 MCQ · 2.0 marks

      In W-GAN, enforcing the 1-Lipschitz condition on the critic (discriminator) is crucial. Which of the following methods **fails** to properly enforce the Lipschitz constraint?
      1. [[IMAGE:ce01f99902ff74dc_6_22]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_6_23]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_6_24]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_6_25]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 200 MCQ · 2.0 marks

      If a domain classifier achieves 100% accuracy in distinguishing between source and target domain samples, what can we say about domain adaptation?
      1. The domains are perfectly aligned
      2. The classifier is overfitting
      3. The domains are misaligned
      4. Nothing can be inferred

      A published solution is not available for this question yet.

      Question 201 MCQ · 2.0 marks

      [[IMAGE:ce01f99902ff74dc_7_26]]
      Source diagram or notation
      1. [[IMAGE:ce01f99902ff74dc_7_27]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_7_28]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_7_29]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_7_30]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 202 MCQ · 3.0 marks

      In the PyTorch implementation of W-GAN, which of the following code snippets correctly enforces the weight clipping constraint on the critic?
      1. [[IMAGE:ce01f99902ff74dc_8_31]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_8_32]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_8_33]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_8_34]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 203 MCQ · 3.0 marks

      [[IMAGE:ce01f99902ff74dc_8_35]]
      Source diagram or notation
      1. [−1, 2]
      2. [1,−2]
      3. [−2, 4]
      4. [0, 0]

      A published solution is not available for this question yet.

      Question 204 MCQ · 4.0 marks

      [[IMAGE:ce01f99902ff74dc_9_36]]
      Source diagram or notation
      1. [[IMAGE:ce01f99902ff74dc_9_37]]
        Source diagram or notation
      2. [[IMAGE:ce01f99902ff74dc_9_38]]
        Source diagram or notation
      3. [[IMAGE:ce01f99902ff74dc_9_39]]
        Source diagram or notation
      4. [[IMAGE:ce01f99902ff74dc_9_40]]
        Source diagram or notation

      A published solution is not available for this question yet.

      Question 205 MCQ · 4.0 marks

      [[IMAGE:ce01f99902ff74dc_10_41]]
      Source diagram or notation
      1. 1.1
      2. 1.5
      3. 0.8
      4. 0.9

      A published solution is not available for this question yet.

      Question 206 NAT · 3.0 marks

      [[IMAGE:ce01f99902ff74dc_10_42]]
      Source diagram or notation

        A published solution is not available for this question yet.

        Question 207 NAT · 2.0 marks

        [[IMAGE:ce01f99902ff74dc_11_43]]
        Source diagram or notation

          A published solution is not available for this question yet.

          Question 208 NAT · 2.0 marks

          [[IMAGE:ce01f99902ff74dc_11_44]]
          Source diagram or notation

            A published solution is not available for this question yet.

            Question 209 MCQ · 3.0 marks

            [[IMAGE:ce01f99902ff74dc_12_45]] Based on the above data, answer the given subquestions.
            If the generator consists of three linear layers: 100 → 128 → 256 → 784, how many trainable parameters are in the generator (excluding biases)?
            Source diagram or notation
            1. 78,400
            2. 102,400
            3. 246,272
            4. 156,000

            A published solution is not available for this question yet.

            Question 210 MCQ · 3.0 marks

            [[IMAGE:ce01f99902ff74dc_12_45]] Based on the above data, answer the given subquestions.
            During Bi-GAN training, suppose you compute discriminator output on: [[IMAGE:ce01f99902ff74dc_12_46]] If using Binary Cross Entropy loss with target labels 1 for real and 0 for fake, what is the total discriminator loss?
            Source diagram or notationSource diagram or notation
            1. 0.324
            2. 0.45
            3. 0.23
            4. 0.85

            A published solution is not available for this question yet.

            Question 211 MCQ · 3.0 marks

            [[IMAGE:ce01f99902ff74dc_12_45]] Based on the above data, answer the given subquestions.
            In a Bi-GAN, you use an encoder E, generator G, and discriminator D. Given batch size 64, and each sample is 784-dimensional, what is the shape of the input to the discriminator?
            Source diagram or notation
            1. [64, 784]
            2. [64, 100]
            3. [64, 884]
            4. [128, 784]       **DLP** **Section Id :** 64065391704 **Section Number :** 11 **Section type :** Online **Mandatory or Optional :** Mandatory **Number of Questions :** 13 **Number of Questions to be attempted :** 13 **Section Marks :** 50 **Display Number Panel :** Yes **Section Negative Marks :** 0 **Group All Questions :** No **Enable Mark as Answered Mark for Review and** No **Clear Response :** **Section Maximum Duration :** 0 **Section Minimum Duration :** 0 **Section Time In :** Minutes **Maximum Instruction Time :** 0

            A published solution is not available for this question yet.