Deep Learning Unit 2 – Deep Networks: AKTU previous year questions
28 AKTU questions from Unit 2 (Deep Networks) asked in 2023–2026, tagged by marks, year and topic. In short: about 38 marks of every paper come from this unit; the most asked topic is Convolutional Networks (9 times); 4 questions came back in a later year. The latest ones are listed below; on the page you can filter them by 2-mark or long questions, topic and repeats.
Unit 2 previous year questions
- Discuss convolutional neural network (CNN)? (2 marks, 2026, Convolutional Networks)
- Name any two activation functions used in deep networks. (2 marks, 2026, Convolutional Networks)
- Name two applications of Word2Vec. (2 marks, 2026, Generative Adversarial Networks (GAN))
- Describe how probabilistic models help reduce overfitting. Explain with examples. (7 marks, 2026, Regularization & Batch Normalization)
- Compare batch normalization with layer normalization and instance normalization. (7 marks, 2026, Regularization & Batch Normalization)
- Explain the concept of semi-supervised learning with examples. Compare it with supervised and unsupervised learning. (7 marks, 2026, Semi-supervised Learning)
- Explain the significance of the “AI Winter” in deep learning history. (2 marks, 2025, History & Theory of Deep Learning)
- Discuss the role of convolutional layers in Convolutional Neural Networks. (2 marks, 2025, Convolutional Networks)
- Trace the historical development of deep learning that enabled deep networks to outperform traditional machine learning models. (2 marks, 2025, History & Theory of Deep Learning)
- How does weight sharing in CNNs improve efficiency and performance? (2 marks, 2025, Convolutional Networks)
- What is the probabilistic theory of deep learning? (2 marks, 2025, History & Theory of Deep Learning)
- Compare deep and shallow networks in terms of representational capacity and computational complexity. (2 marks, 2025, History & Theory of Deep Learning)
Most asked Unit 2 topics
- Convolutional Networks – asked 9 times
- History & Theory of Deep Learning – asked 8 times
- Regularization & Batch Normalization – asked 4 times
Unit 2 question pattern
- 2-mark questions: 11 asked in 2023–2026
- 7-mark questions: 3 asked in 2023–2026
- 10-mark questions: 14 asked in 2023–2026
- About 38 marks from Unit 2 in every paper
- 4 questions were asked again in a later year
Unit 2 questions year by year
- 2026: 6 Unit 2 questions asked (27 marks across that year's papers)
- 2025: 15 Unit 2 questions asked (102 marks across that year's papers)
- 2023: 7 Unit 2 questions asked (54 marks across that year's papers)