Skip to content

Deep Learning (DL) important questionsMost repeated in SPPU papers

These 13 questions appeared in more than one SPPU Deep Learning paper we have: 11 papers, from Oct 2022 to Nov 2025. The most-asked come first. Each one links to the papers it was asked in.

All DL papers and notes

Repeated questions

Endsem questions (12)

  1. 4×asked

    What is Recurrent Neural Network (RNN)? Draw and Explain Simple Recurrent Neural Network. Explain four types of RNN with proper diagram.

    Nov 2024 ·May 2024 ·Nov 2023 ·Nov 2022

  2. 4×asked

    Write a short note on Long Short-Term Memory Networks (LSTM).

    May 2024 ·Nov 2023 ·May 2023 ·Nov 2022

  3. 3×asked

    Explain deep learning-based recommender systems with suitable diagrams.

    Nov 2025 ·May 2025 ·May 2023

  4. 3×asked

    How is social network analysis done using deep learning architectures? 

    Nov 2025 ·May 2025 ·May 2023

  5. 3×asked

    What is greedy layer-wise pre-training network?

    Nov 2025 ·May 2025 ·Nov 2023

  6. 3×asked

    Explain content based, collaborative and hybrid recommender system with pros and cons.

    Nov 2024 ·May 2024 ·Nov 2022

  7. 3×asked

    Explain how sequence to sequence model works.

    May 2024 ·May 2023 ·Nov 2022

  8. 2×asked

    What is a regularized auto-encoder? Where is it used?

    Nov 2025 ·May 2025

  9. 2×asked

    Differentiate between Undercomplete Autoencoder and Sparse Autoencoder.

    Nov 2024 ·May 2024

  10. 2×asked

    Explain Transfer Learning and Domain Adaption.

    Nov 2023 ·Nov 2022

  11. 2×asked

    What are the application areas of image classification? Explain CNN for image Classification. 

    May 2023 ·Nov 2022

  12. 2×asked

    Why is the network called Greedy Layer Wise Pretraining Network?

    May 2023 ·Nov 2022

Insem questions (1)

  1. 2×asked

    What is the problem of vanishing gradient? Explain the various solutions to avoid it? 2

    Sep 2024 ·Oct 2022

Questions count as repeated when their wording matches closely across papers. Numericals that depend on a figure are left out, since the figure changes from paper to paper.