Statistics · Optimization · AI

Spandan Roy

Doctoral Student in Decision Sciences at IIM Bangalore, working at the intersection of statistical learning, optimization, risk-aware modelling, and decision-making under uncertainty.

About

Research profile and academic background.

I am a Doctoral Student in Decision Sciences at Indian Institute of Management Bangalore. My academic training is in Statistics, with a Bachelor's and Master's degree from Hindu College, University of Delhi.

My interests lie at the intersection of statistical learning, optimization, and decision-making under uncertainty. I am particularly interested in risk-averse modelling, distributionally robust optimization, extreme-value and tail-risk methods, causal inference, and statistical foundations of AI.

I also work on applied machine learning and deep learning implementations, primarily as a way to connect statistical methodology with reproducible computational practice.

  • Statistical learning and asymptotics
  • Distributionally robust optimization
  • Extreme-value and tail-risk modelling
  • Decision-making under uncertainty
  • AI audit and fairness in hiring systems

Research

Broad areas of work.

Statistical Evaluation of Algorithmic Decision Systems

Algorithmic fairness · LLM evaluation · Computational social science · Advised by Prof. Subhabrata Majumdar

This work studies large language models as decision aids in hiring pipelines. Using controlled synthetic-resume experiments, the project examines whether LLMs exhibit differential shortlisting behaviour across region, gender, institutional prestige, and caste-associated name signals in the Indian hiring context. The broader goal is to understand when apparently merit-based automated evaluations become sensitive to social identity signals, prompt framing, or incomplete candidate evidence.

Semi-parametric Extreme Quantile Estimation

Extreme-value theory · Non-parametric statistics · Machine learning · Advised by Prof. Soudeep Deb

This work focuses on estimating very high conditional quantiles and tail risks when ordinary data are sparse in the extreme region. The methodological emphasis is on combining non-parametric statistical ideas, extreme-value theory, and machine-learning-based function approximation to study rare-event regimes where standard prediction methods can fail or provide weak uncertainty guarantees.

Rare-event and Systemic-risk Modelling under Partial Information

RST problem · Large deviations · Systemic risk · Partial information · Advised by Prof. Anand Deo

This line of work is concerned with rare systemic-risk events in networked financial systems, especially when the full agent-asset allocation structure is only partially observed. The problem connects large-deviation ideas, partial-information modelling, and stress-testing logic, with the aim of characterizing how extreme losses can emerge from incomplete knowledge of the underlying network.

Importance Sampling for Tail-risk and Rare-event Estimation

Rare-event simulation · Variance reduction · Coherent risk measures

This project studies importance sampling methods for estimating rare-event probabilities and tail-sensitive risk functionals such as VaR, CVaR, expected shortfall, and distortion-based risk measures. The broader aim is to design simulation procedures that improve estimation accuracy in the tail, where naive Monte Carlo methods often require prohibitively large sample sizes.

Selected Work

Technical and applied projects.

Applied Computer Vision Implementations

Technical implementation / computer vision

A collection of OpenCV-based computer vision implementations covering image processing, feature extraction, object or scene analysis, and core computer vision workflows.

Image Classification and Segmentation

Applied deep learning project

Implementation of deep learning workflows for image classification and segmentation tasks, focusing on data preparation, training, evaluation, and reproducibility.

Virtual Assistant and Sentiment Analysis

Applied NLP project

A prototype virtual assistant with sentiment-aware response behaviour, combining basic NLP, sentiment classification, and rule-based assistant logic.

Publications

Papers and manuscripts.

Current manuscripts in progress.

Google Scholar

Writing

Academic notes and travel essays.

Academic Notes

Statistics / OR / Mathematics / Machine Learning

Short expository posts introducing topics I am studying or using in research. Longer derivations, proofs, and examples can be linked as PDF notes.

  • Probability Theory
  • Linear Programming
  • Extreme Value Theory
View Academic Notes

Travel & Culture

Travel essays / photo notes / cultural observations

Essays and photo notes from places I visit, with attention to local culture, food, landscapes, people, and the small details that shape how a place is experienced.

  • City walks and local histories
  • Food, markets, and everyday life
  • Mountains, beaches, and landscapes
  • Photography-led travel reflections
View Travel Writing