Sayan Ranu

portrait

News

ICLR hattrick!:

We will be presenting three papers spanning graph distillation, GNN explainers and learning brownian dynamics using GNNs.

Paper Alert!:

Generative AI for graphs has been a vibrant area of research. Yet, little attention has been paid to generative modeling under data scarcity. Not any more! Introducing GShot, which will appear in LOG 2023.

NeurIPS 2023!:

Our work on tracking pollution data, led by Rijurekha Sen, will appear at NeurIPS 2023!

Sept, 2023:

Honored and delighted to be recognized as an Indian National Science Academy (INSA) Associate Fellow.

June, 2023:

Delighted to receive Google research scholar award 2023 with Anoop for "Reinforcement Learning on Rough Landscapes for Materials Discovery". Thank you Google! We are the only team from India to receive the award this year.

May, 2023: Delighted to have a presence in KDD!

Our work on frugal spatio-temporal forecasting on road networks will appear in KDD2023. We take a practical approach that is cognizant of the challenges in sensor maintenance and noise in their readings. Congrats Mridul!

April, 2023: 2-out-of-2 at ICML!

Delighted that both of our ICML submissions have been accepted. In the first work, we develop a GNN architecture to handle dynamic feature sets. In the second work, we solve a combinatorial optimization problem on graphs using GNN+RL to design better materials.

March, 2023: Best-of-WSDM2023-selection!

Delighted that our work on GNN explainability has been recognized as among the top-10 papers of WSDM 2023.

Paper Alert!:

Our collaboration with Flipkart on fair allocation in social e-commerce accepted to WWW 2023.

Jan, 2023:

Honored and humbled to receive Mrs. Veena Arora Early Career Faculty Research Award of IIT Delhi for the year 2021-22. Snippet from the ceremony.

Jan, 2023: Delighted to have a presence in ICLR!

In this work, we show that graph neural ODEs with appropriate inductive bias can learn the dynamics from small data and generalise it to unseen system sizes.

Jan, 2023:

Both AAAI papers from the group selected for oral presentations.

Jan, 2023:

Our paper on global counter-factual analysis has been selected for oral presentation in WSDM 2023.

Dec, 2022: Double delight at AAAI, again!

For two years in a row, we will be presenting two papers at AAAI. In the first paper, we learn to solve Mixed Integer Programs under continual flow of training data. In the second paper, we show GNNs are vulnerable to adversarial attacks even if the attacker is unaware of the model parameters, model type, or the task being solved by the GNN. Congrats to Sahil, Kartik and Samidha!

I hold the joint positions of Nick McKeown Chair Associate Professor at the Department of Computer Science and Eng. and the Yardi School of AI at IIT Delhi. My research interests span the broad area of network science with particular focus on graph neural networks and spatio-temporal data analytics. The tag-cloud of my paper abstracts from last five years provides an accurate representation of my current research interests.

I graduated from the Computer Science department of University of California, Santa Barbara in March, 2012 under the supervision of Prof. Ambuj Singh. Prior to joining IIT D, I worked as Assistant Professor at IIT Madras for close to three years and at IBM Research, Bangalore for a year and a half.

Internships: I am NOT offering any short-term projects. My apologies if I don't respond to your email.

Contact:

  • Office: Bharti 421
  • sayanranu@gmail.com
  • sayanranu@cse.iitd.ac.in
  • Twitter
  • Linked In