Hi — I'm Viral Chitlangia. And, this is me.
I am a statistics student, and have worked on MCMC methods, Generative Models for Spatial-Temporal Data, Regression Models among other stats stuff. I like stats, cause it is fun, and I build models, cause that is my life.
About Me
I am currently an Undergraduate Student in Statistics & Data Science in IIT Kanpur. I have worked on various aspects of Statistics and Interdisciplinary fields. I have had the pleasure of working on various projects with Professor Sharmishtha Mitra, Professor Dootika Vats, Professor Suncica Hadzidedic and Professor Swapnil Mishra. I have worked on various topics on statistics, including Regression Modelling, Markov Chain Monte Carlo and Bayesian Modelling. I have worked on Generative models in my projects, including Variational Autoencoders. My projects, have also gotten me to work in interdisciplinary fields, like Public Health, and my project on Digital Interventions for Loneliness.
Education
B.S. Statistics & Data Science — IIT Kanpur (2022 — 2026)
CPI: 9.0/10.0 — Minor: CS (Machine Learning)
Publications & Preprints
-
View
Swap Regression Methodology for Predicting Relationship with Historical Bivariate Data
Viral Chitlangia, Mosuk Chow, Sharmishtha Mitra — arXiv:2508.15479
Major Projects
Swap Regression
- Analyzed the paper by Mosuk Chow, Bing Li, and Jackie Q. Xue, ON REGRESSION FOR SAMPLES WITH ALTERNATING PREDICTORS AND ITS APPLICATION TO PSYCHROMETRIC CHARTS, and developed models with alternating predictors on bivariate data using SWAP Regression.
- Defined a new loss function for the Swap Regression model.
- Implemented ALT-OPT to solve the loss function.
- Tested the model on real data: US Public Debt and GDP.
- Applied the model to predict the causality direction of US Public Debt and GDP without prior knowledge.
Digital Intervention for Loneliness
- Explored Reddit and Google to find relevant subreddits and apps which target loneliness.
- Scraped Reddit, Google Play, Apple Play, and A Lonely Life (a loneliness forum), using Python libraries and Rvest (R) to collect text data on loneliness.
- Applied Topic Modelling to cluster the data into relevant topics, identifying what people are talking about online and finding areas where apps could be improved.
Deep Generative Models for Spatial-Temporal Data
- Implemented and analyzed the paper on AggVAE written by Swapnil Mishra et al.
- Worked on combining AggVAE with Population Disaggregation techniques to improve predictions using low-resolution data points.
- Applied the model on US COVID Data, segregated by 9 regions of the country.
Enveloping Techniques for Importance Sampling in MCMC
- Studied the paper by Apartim Shukla, Dootika Vats, and Eric C. Chi, MCMC Importance Sampling via Moreau-Yosida Envelopes.
- Explored various enveloping techniques in optimization.
- Analyzed the properties of different envelopes and their usability in sampling, especially Importance Sampling.
MCMC Machine Unlearning
- Understood and implemented the paper Markov Chain Monte Carlo-based Machine Unlearning: Unlearning What Needs to Be Forgotten by QP Nguyen et al.
- Implemented a novel algorithm for Machine Unlearning Sampling using Newton's Method Update, inspired by Certified Data Removal from Machine Learning Models by C. Guo et al.
- Compared the algorithm with other prominent MCMC sampling algorithms on Logistic and Negative Binomial regression data.
Random Projects
Get in touch
Feel free to mail on either mail ids. Click on any button below to copy to clipboard.