Publications

• Aatif Nisar Dar, Nandana Sengupta, and Chetan Arora, “Assessing the Feasibility and Ethics of Economic Status Prediction using Deep Learning on Household Images.” ACM Journal on Computing and Sustainable Societies (2024).

Code: Link to the code
paper: Link to the Paper

• Aatif Nisar Dar and Reshma Rastogi, “MLGAN: Addressing Imbalance in Multilabel Learning Using Generative Adversarial Networks,” 2023 International Conference on Emerging Techniques in Computational Intelligence (ICETCI), Hyderabad, India, 2023, pp. 324-331, doi: 10.1109/ICETCI58599.2023.10331105. - Best Presenter Award

Code: Link to the code
Paper: Link to the Paper

• Aatif Nisar Dar, “Principal Component Analysis,” Global Scientific Journal, GSJ: Volume 9, Issue 7, July 2021, Online: ISSN 2320-9186. (Lightly Peer-reviewed)

Paper: Link to the Paper

Posters/Workshops/Symposiums

• Aatif Nisar Dar, Badrinath Padmanabhan, Mrunal Atul Kadane, Aaditeshwar Seth. “Using satellite images to track relative socio-economic development in rural India”. Poster presented at ACM COMPASS ‘24, New Delhi, India.

Code: Link to the code
paper: Link to the Paper

• Research Week with Google 2024. (Funded travel) 01 - 03 Feb ‘24
• Symposium on Machine Learning, IIT - Bombay. (Funded travel) 21 - 23 Dec ‘23

Other Writings

• Aatif Nisar Dar, “How to Choose the Best Classification Model,” Medium, Oct ‘22.

• Aatif Nisar Dar, “Partial Multi-Label GANs,” Medium, Oct ‘21.

• Aatif Nisar Dar, “Image-To-Image Translation via Generative Adversarial Networks (GANs),” Medium, Dec ‘21.

Projects

Research Projects

Tracking Socio-Economic Development in Rural India over Two Decades Using Satellite Imagery (Jan 2024 - Present)

• Examining the evolution of rural India’s socio-economic landscape across a span of two decades through the analysis of satellite imagery and nightlight data.
• This work provides evidence of the possible impact that policy changes and other factors have on the development of regions.

Code: Link to the code

Master’s Dissertation

Generative Adversarial Networks and its Applications, South Asian University. (May 2021 - May 2022)

• Executed GAN on MNIST Digit dataset and MNIST Fashion dataset with Self Attention Module and Spectral Normalization in both Generator and Discriminator. Added TTUR (Two Time Scale Update) to stabilize the training of the GAN.
• Implemented CycleGAN, Pix2PixGAN, StyleGAN, SMIT, and AttentionGAN on the CelebA dataset.
• Outcome of my dissertation was a novel architecture called MLGAN (Multi-Label Generative Adversarial Network), which was designed to address the challenge of data imbalance in multilabel datasets.

Code: Link to the code
Dissertation: Link to the dissertation

Other Projects

• PHP
  Designed a website ‘Learn PHP’ for my semester project using HTML, CSS, and PHP with Apache as the webserver. (2019)
  GitHub Repository

• Software Engineering Project
  Developed an online web-based tuition platform where students get help in finding suitable teachers to study a particular subject. (2019)
  GitHub Repository

• Android Studio
  Created an online attendance apk using Android Studio where the teacher enters student details at once and marks students either present or absent. (2018)
  GitHub Repository