Deep Learning Unit 3 – Dimensionality Reduction: important questions for AKTU
Unit 3 (Dimensionality Reduction) questions that AKTU repeats most often. This unit carries about 26 marks per paper. Start with the repeated questions, then the most asked topics.
Most repeated Unit 3 questions
- Compare PCA and LDA. Explain how LDA maximizes class separability using scatter matrices. (7 marks, 2026, Linear Dimensionality Reduction (PCA, LDA))
- Explain convolution operation mathematically, including stride, padding, and feature maps. (7 marks, 2026, ConvNet Architectures (AlexNet, VGG, ResNet))
- Compare AlexNet, VGG, Inception, and ResNet in terms of depth, performance, and computation. (7 marks, 2026, ConvNet Architectures (AlexNet, VGG, ResNet))
Most important Unit 3 topic
- Linear Dimensionality Reduction (PCA, LDA) (Unit 3: Dimensionality Reduction) – asked 7 times in 2023, 2025, 2026
Most asked Unit 3 topics
- Linear Dimensionality Reduction (PCA, LDA) – asked 7 times
- ConvNet Architectures (AlexNet, VGG, ResNet) – asked 6 times
- Training a ConvNet (Weight Init, Hyperparameters) – asked 5 times
More Unit 3 previous year questions
- Describe the role of regularization during optimization. (2 marks, 2026, Training a ConvNet (Weight Init, Hyperparameters))
- Explain the impact of weight initialization on the performance of deep networks. (2 marks, 2025, Training a ConvNet (Weight Init, Hyperparameters))
- Define Linear Discriminant Analysis (LDA). (2 marks, 2025, Linear Dimensionality Reduction (PCA, LDA))
- Discuss how batch normalization impacts the training of convolutional neural networks. (2 marks, 2025, Training a ConvNet (Weight Init, Hyperparameters))
- How autoencoders can be used to learn low-dimensional representations of data? (2 marks, 2025, Autoencoders)
Unit 3 syllabus topics
- Linear Dimensionality Reduction (PCA, LDA)
- Autoencoders
- ConvNet Architectures (AlexNet, VGG, ResNet)
- Training a ConvNet (Weight Init, Hyperparameters)