AGNIDEEP “AGNI” AICH

PhD Candidate · Statistical Machine Learning

"To truly laugh, you must be able to take your pain, and play with it." — Charlie Chaplin
About

My Story

Agnideep Aich

I am a PhD candidate in Statistical Machine Learning at the University of Louisiana at Lafayette, working under Dr. Bruce Wade . I combine statistics and machine learning, with a focus on dependence modeling using copulas, to support risk prediction and analysis in population health and biomedical datasets, including large-scale public health surveys and clinical datasets.

Outside the lab, I live loudly. I collect sunglasses and colognes, spend too much time thinking about cars, and I am an Arsenal fan (even though lately, let’s not talk about them). I’m also into EDM (trance, future bass, and synthwave), and I’m a huge MMA fan. Also, I love eating, and if I see a Raising Cane’s, my heart starts beating a little faster.

Focus Areas

Research Interests

Statistical Machine Learning

I combine statistical theory with machine learning to build reliable prediction methods and evaluate them rigorously on real datasets. My work spans both methodological research and theory-driven results, including fully theoretical projects in statistical diagnostics.

Copulas

Copulas model dependence beyond Gaussian assumptions by separating marginal behavior from joint structure. I use copulas to capture non-linear and tail dependence that commonly appears in population health and biomedical data, and to build dependence-aware learning tools.

Risk Prediction

Risk prediction is a central theme in my work, especially for health outcomes. I develop and study models that improve prediction performance when features are dependent, with applications to large-scale public health and clinical datasets.

Population Health

I work with population health datasets that include cross-sectional surveys and irregular longitudinal measurements. These settings often involve dependence, missingness, and heterogeneity, and my methods are designed with those realities in mind.

AI in Healthcare

I am interested in turning AI methods into tools that are reliable for healthcare settings. My focus is on developing models that respect dependence structure in the data and produce predictions that remain stable across diverse patient populations and real-world conditions.

Experience

Work Experience

August 2025 – December 2025

Statistical and Data Science Research Assistant

University of Louisiana at Lafayette

I analyzed public health datasets using statistical modeling, machine learning, and data visualization under Dr. Amanda Mayeaux . I also collaborated across public health, education, and kinesiology to frame research questions and design analyses, contributing to manuscripts in preparation.

August 2024 – May 2025

Research Assistant

Statistical Consulting Center, University of Louisiana at Lafayette

Provided statistical consulting, data analysis, and data visualization support to PhD students, faculty, and other clients. Collaborated closely with researchers to translate study objectives into statistical questions and deliver robust, actionable solutions.

August 2020 – May 2024

Instructor of Record

University of Louisiana at Lafayette

Taught two sections each semester of Elementary Statistics (STAT 214) and Applied College Algebra (MATH 105). Developed course materials including lectures, assignments, and exams, and received strong student feedback for clear communication and engagement.

Education

Education

2021 – May 2026 (Anticipated)

Ph.D. in Statistical Machine Learning

University of Louisiana at Lafayette

Dissertation: Learning from Extremes: A Copula-Based Feature Selection Framework for Machine Learning-Driven Risk Prediction.

2019 – 2021

M.S. in Mathematics

University of Louisiana at Lafayette

2016 – 2018

M.S. in Statistics

University of Calcutta, India

2013 – 2016

B.S. in Statistics

Asutosh College, India

Minors: Mathematics & Computer Science.

Recent Work

Recent Publications

2026

Bayesian Inference for Joint Tail Risk in Paired Biomarkers via Archimedean Copulas with Restricted Jeffreys Priors

Agnideep Aich, Md. Monzur Murshed, Sameera Hewage, Ashit Baran Aich

Preprint

DOI

2025

A Copula Based Supervised Filter for Feature Selection in Machine Learning Driven Diabetes Risk Prediction

Agnideep Aich, Md. Monzur Murshed, Sameera Hewage, Amanda Mayeaux

Preprint

DOI

2026

The Minimax Lower Bound of Kernel Stein Discrepancy Estimation

J. Cribeiro-Ramallo, Agnideep Aich, F. Kalinke, Ashit Baran Aich, Zoltán Szabó

Accepted at AISTATS 2026 (Preprint)

DOI
Get In Touch

Contact

I’m always open to research collaborations, academic discussions, or just a good conversation. If you want to talk statistics, machine learning, or life outside research, reach out. Email me, or DM me on Instagram. I always respond!!