Kartik Sharma

searching for the blood flow...

prof_pic.jpg

☍ 🤖 🔎 🦺 🎛️

I am a third-year Ph.D. student at Georgia Institute of Technology, being advised by Prof. Srijan Kumar and mentored by Dr. Rakshit Trivedi. I have had the pleasure of spending the summer of 2023 at Visa Research. Before this, I graduated from IIT Delhi in 2021, where I was fortunate to be advised by Prof. Sayan Ranu and mentored by Prof. Sourav Medya.

Through my research, my goal is to study and build machine learning (ML 🤖) models for structured data (specifically, graphs ☍), which are reliable and can be easily controlled by the users. To this end, I am focused at measuring the reliability and alignment of existing models (🔎), designing structure-inspired robust (🦺) and user-controllable models (🎛️). During my Ph.D. journey so far, I have explored specific research questions within these areas and am excited to continue my exploration in the future.

news

May 14, 2024 Spending my summer at Microsoft Research, Redmond in the Research for Industry Lab.
May 02, 2024 Paper accepted at ICML 2024 (paper and code coming soon!). See you in Vienna!
Apr 15, 2024 Survey on social recommendation accepted at ACM CSUR 2024 (link).
Mar 07, 2024 Website updated.

selected publications

Measuring ML reliability and alignment for structured data.

  1. Task and model agnostic adversarial attack on graph neural networks
    Kartik Sharma, Samidha Verma , Sourav Medya , and 2 more authors
    In Proceedings of the AAAI Conference on Artificial Intelligence , 2023
  2. Temporal dynamics-aware adversarial attacks on discrete-time dynamic graph models
    Kartik Sharma, Rakshit Trivedi , Rohit Sridhar , and 1 more author
    In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , 2023

Designing reliable and interpretable predictive models.

  1. Representation learning in continuous-time dynamic signed networks
    Kartik Sharma, Mohit Raghavendra , Yeon-Chang Lee , and 1 more author
    In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management , 2023

Enabling user-controllable generation.

  1. Diffuse, Sample, Project: Plug-And-Play Controllable Graph Generation
    Kartik Sharma, Srijan Kumar , and Rakshit Trivedi
    In Proceedings of the 41st International Conference on Machine Learning (ICML) , 2024