About Me

I’m a Fourth Year Undergraduate at the Indian Institute of Technology Kanpur (IITK), pursuing a BTech. in Electrical Engineering with a Double Major in Mathematics. My research interests lie in the mathematical foundations of Machine Learning, with a focus on Optimization, High Dimensional Statistics, and Approximate Inference. I also maintain an active interest in continuous-time Machine Learning with ODEs and SDEs, and Deep Generative Models.

Since my Sophomore Year, I’ve been working as a Research Assistant in Prof. Piyush Rai’s group at IITK, focusing on Approximate Inference and Deep Generative Models. My previous work under Prof. Rai’s group involved developing Generative Models for Joint Image and Video Generation, and nonparametric priors for GANs using Second-Order ODEs. Currently, I am working on developing an extension of the Model Agnostic Meta-Learning (MAML) Algorithm for meta-learning of continuous-time dynamical systems, and investigating results from SDE theory to improve the performance of Stochastic Gradient MCMC Algorithms.

I spent my Third Year as an Exchange Student at Aalto University School of Science, Finland, where I worked as a Research Assistant under Dr. Markus Heinonen and Prof. Harri Lahdesmaki, focusing on Approximate Bayesian Inference and Adversarial Learning for continuous-time generative models governed by ODEs and SDEs.

I’m currently looking for Research Assistantship and collaboration opportunities in Optimization, Approximate Inference, High Dimensional Statistics, and Learning Theory. If you’re interested in a collaboration or have an opening in your research group where I could be of some use, please feel free to contact me.


  • Optimization
  • High Dimensional Statistics
  • Approximate Inference
  • Learning Theory
  • Gradient Based MCMC
  • Stein Variational Gradient Descent
  • Continuous Time ML with ODEs and SDEs
  • Deep Generative Models


  • Bachelor's Exchange Student, Jan' 2020 - Jan 2021

    Aalto University School of Science, Finland

  • BTech in Electrical Engineering and BS in Mathematics (Double Major), 2017-2022 (expected)

    Indian Institute of Technology Kanpur, India


NeurInt : Learning to Interpolate through Neural ODEs

We propose a generative model for images that learns a distribution of continuous-time interpolation trajectories for a given …

Joint Image and Video Generation using Residual Vectors

We develop a hierarchical probabilistic model for joint image and video generation that generates a summary frame for the video, and …

TorchGAN: A Flexible Framework for GAN Training and Evaluation

TorchGAN is a PyTorch based framework for writing succinct and comprehensible code for training and evaluation of Generative …



Continuous-Discrete Bayesian Filtering for Stochastic Differential Equations

Supervisor: Prof. Simo Särkkä, Aalto University As part of the term project for Aalto University’s course on State Space Models, …

Nonparametric Models for Image Generation and Interpolation

Supervisor: Prof. Piyush Rai, IIT Kanpur We developed a generative model for images that uses Probabilistic Second Order Neural ODEs to …

Probabilistic Models for Joint Image and Video Generation

Supervisors: Prof. Piyush Rai and Prof. Vinay Namboodiri, IIT Kanpur We developed a hierarchical probabilistic model for joint image …

TorchGAN: A Flexible Framework for GAN Training and Evaluation

I co-developed TorchGAN, a lightweight customizable PyTorch framework for training and evaluation of Generative Adversarial Networks. …

Computational Models for Human Inference of Social Situations

Supervisor: Prof. Nisheeth Srivastava, IIT Kanpur

The aim of the project was to characterize the nature and granularity of information …

Attention Based Models for Image Captioning and Generation

As part of a summer project offered by Programming Club, IIT Kanpur, I studied the theory of encoder decoder sequence models, soft …

Case Studies : A Legal Consultant Bot

I developed a legal consultant bot that summarises a legal document, highlights the key phrases and answers some rudimentary questions …