Research Interests

Wasing the always of wanting of knowing.
The eternal desire of a hungry soul is knowledge.
                                                            – Mistborn Series, Brandon Sanderson

My goal is to use theoretical and empirical insights to develop a principled understanding of small-scale representations of graphs and machine learning (ML)models so that they are reliable when deployed in the real-world. In particular, my focus is on Node Embeddings which distill high-dimensional feature and neighbourhood information onto low-dimensional spaces. Currently, I am thinking about the following problems:

  • Node embedding algorithms are typically designed for ML tasks such as node classification and link prediction. How applicable are these embeddings for traditional graph mining tasks such as summarization and query answering? How do they compare against other types of small-scale representations?

  • Node embeddings are smaller abstractions of graphs which model users and relationships in the context of social networks. Do such models exacerbate or reduce the harmful biases of real-world applications? How do we account for bias in the data so that it does not propagate into the embedding?

In the past, I worked on a related theme in the context of machine teaching. How can we construct useful training examples to simultaneously train stochastic gradient descent learners when we have incomplete or noisy information about them?

On the application side, I am excited about computational social science wherein I am interested in human behaviour on online social networks. My past research has explored how platforms facilitate trustworthiness amongst users through incentives and gamification. I have also studied the role of digital signalling on the actions, social capital, and social status of users online.

Experience

 
 
 
 
 
January 2019 – Present
University of Helsinki

Graduate Research Assistant

with Michael Mathioudakis

Responsibilities include

  • Developing theory, algorithms, and applications of low-dimensional representations of graphs, specifically focusing on node embeddings.
  • Using machine learning, deep learning and graph theoretic methods to quantify how brittle algorithms may be to random as well as structural perturbations, uncertainty in the form of noisy or unavailable data, and temporal changes at appropriate levels of granularity.
  • Designing novel techniques with theoretical and empirical guarantees, and useful applications towards downstream tasks in mining social and biological networks.
 
 
 
 
 
July 2021 – November 2021
UPF Barcelona

Doctoral Intern

with Carlos Castillo

Responsibilities include

  • Understanding algorithmic fairness of machine learning algorithms on graphs, particularly node embeddings and graph neural networks in the context of node classification.
  • Analyzing philosophical underpinnings of representing attributed graphs in low dimensions.
 
 
 
 
 
March 2018 – November 2018
IIIT-Delhi

Research Associate

with Ponnurangam Kumaraguru

Responsibilities include

  • Studied the role of game elements such as badges and reputation points in characterizing underlying social qualities like popularity and impact of users on Stack Overflow.
  • Applied statistical learning, and time-series analysis methods to empirically quantify and validate the strength of game elements/digital signals on a rich dataset of 3,831,147 users and their activities spanning over a decade.
  • Analyzed the presence of costly to earn and hard to observe signals to qualitatively differen-tiate between highly impactful and highly popular users.
 
 
 
 
 
August 2017 – February 2018
MPI-SWS

Visiting Research Scholar

with Adish Singla

Responsibilities include

  • Designing novel, state-of-the-art algorithms for teaching a classroom of online projected gradient-descent learners with provable guarantees under complete and incomplete information paradigms.
  • Studied applications of machine teaching methods on synthetic and real-world data for binary classification and handwriting improvement tasks.
  • Developed Bayesian inference algorithms for detecting fake news in social networks and jointly learning users’ flagging accuracy over time.
 
 
 
 
 
August 2016 – June 2017
IIT-Gandhinagar

Junior Research Fellow

with Anirban Dasgupta

Responsibilities include

  • Designed versatile algorithms and novel, unbiased estimators for degree distribution and degree-wise clustering coefficients of large graphs.
  • Conducted experimental evaluations on public network datasets to obtain close estimates of the actual values, for storage less than 1% of the input graph size.
 
 
 
 
 
May 2016 – July 2016
SageMath

Google Summer of Code (GSoC) Intern

with Johan Rosenkilde and David Lucas

Responsibilities include

  • Designed computationally efficient methods for atomic representations of elements of skew polynomial rings and for answering basic questions about them.
  • Implemented methods for computing combinatorial properties, and for encoding and decoding algorithms for Golay, Gabidulin and Rank-Metric Codes.
 
 
 
 
 
May 2013 – July 2013
TIFR, India

Undergraduate Research Intern

with Prahladh Harsha

Responsibilities include

  • Analyzed randomized variants of the deterministic multiple weights update algorithm to show that the total expected cost is only slightly worse than the best possible strategy in hindsight.
  • Surveyed its application to the multi-commodity packing flows problem and a high quality approximation efficiently with provable guarantees.