News


  • [June 2021] Our extended abstract titled "Towards Achieving Adversarial Robustness Beyond Perceptual Limits" has been accepted at the ICML 2021 Workshop on Adversarial Machine Learning. [pdf] [video]

  • [June 2021] Excited to be a recipient of the Qualcomm Innovation Fellowship 2021. This has been awarded to 4 teams in the field of Machine Learning and 13 teams overall in the country. Thankful to Qualcomm for supporting our research on "Efficient Self-Supervised Learning of Robust Representations" for 1 year.

  • [Sept 2020] Thrilled to be a recipient of the Google PhD Fellowship in the area of Machine Learning! This is awarded to 4 students every year in the country. Thankful to Google for supporting my PhD!

  • [Sept 2020] Our paper titled "Guided Adversarial Attack for Evaluating and Enhancing Adversarial Defenses" is accepted for a Spotlight presentation (top 4%) at NeurIPS 2020 [in Proceedings] [video] [arXiv] [code]

  • [May 2020] Thankful for the DeepMind Travel Grant Award at the CVPR 2020 Workshop on Adversarial Machine Learning in Computer Vision

  • [May 2020] Our paper titled "Saliency driven Class Impressions for Feature Visualization of Deep Neural Networks" is accepted at ICIP 2020 [Paper] [arXiv]

  • [April 2020] Thankful for the Google Travel Grant Award for our work accepted at CVPR 2020 conference.

  • [April 2020] Our paper titled "Towards Achieving Adversarial Robustness by Enforcing Feature Consistency Across Bit Planes" is accepted for an Oral presentation at CVPR 2020 Workshop on Adversarial Machine Learning in Computer Vision [pdf] [video] [arXiv] [Code]

  • [Feb 2020] Our paper titled "Towards Achieving Adversarial Robustness by Enforcing Feature Consistency Across Bit Planes" is accepted at CVPR 2020 [pdf] [supp] [video] [arXiv] [Code]

  • [Dec 2019] Thankful for the Pratiksha Travel fellowship for our work accepted at AAAI-2020. Excited to attend my first ML conference!

  • [Nov 2019] Our paper titled "DeGAN : Data-enriching GAN for retrieving Representative Samples from a Trained Classifier", is accepted at AAAI 2020. [Paper] [arXiv] [Code]