Aaradhya Pandey

I am currently a postdoctoral fellow at the Harvard Statistics Department, working with Morgane Austern. Previously, I was a Graduate student at Princeton ORFE and worked with Sanjeev Kulkarni and Arian Maleki. My research interests are at the interface of probability theory, mathematical statistics, and information theory. I am excited to apply the above tools to better understand (classical and quantum) differential privacy, machine unlearning, and spin glasses.

I am trying to develop a mathematical framework to unify Information, Computation, and Quantization.

1. Statistical unlearning of distributions: a hypothesis testing approach
   Aaradhya Pandey, Sanjeev Kulkarni
   Slides arXiv Talk
2. Infinitely divisible privacy and beyond I : Resolution of the \(s^2=2k\) conjecture
   Aaradhya Pandey, Arian Maleki, Sanjeev Kulkarni
   Slides arXiv Talk
3. Gaussian certified unlearning in high dimensions: A hypothesis testing approach
   Aaradhya Pandey, Arnab Auddy, Haolin Zou, Arian Maleki, Sanjeev Kulkarni
   International Conference on Learning Representations (ICLR 2026), Oral
   Slides ICLR arXiv Talk
4. Exact recovery in Gaussian-weighted stochastic block model: information-theoretic thresholds
   Aaradhya Pandey, Sanjeev Kulkarni
   arXiv:2402.12515, 2024 (submitted)
   Slides arXiv
5. Community detection in the hypergraph stochastic block model and reconstruction on hypertrees
   Yuzhou Gu, Aaradhya Pandey
   Proceedings of the 37th Conference on Learning Theory (COLT 2024), PMLR 247:2166–2203, 30 Jun–03 Jul 2024.
   PMLR arXiv Talk

1. Quantum f-differential privacy: a hypothesis testing approach
   Aaradhya Pandey, Morgane Austern
   (In preparation)
   Slides arXiv
2. Quantum infinitely divisible states: a genuinely quantum phenomenon
   Aaradhya Pandey, Morgane Austern
   (In preparation)
   Slides arXiv
3. Multivariate version of the Ghirlanda–Guerra identities and the matrix of spin correlations
   Aaradhya Pandey
   (In preparation)
   Slides arXiv

• [August 2026] Defended my thesis on Differential Privacy and Machine Unlearning at Princeton.
• [Summer 2026] Presented my work on privacy at One World Minds, AIM Caltech, ICERM Brown, and Harvard.
• [March 2026] Presented my work on Infinitely divisible privacy at Columbia University.
• [March 2026] Presented my work on Infinitely divisible privacy at IST, Austria.
• [Feb 2026] Presented my work on Infinitely divisible privacy at IISc, Bangalore, India.
• [Jan 2026] Presented my work on Infinitely divisible privacy at JHU, Baltimore, US.
• [Jan 2026] Presented my work on Infinitely divisible privacy at Google, NYC, US.
• [Jan 2026] Presented my work on Infinitely divisible privacy at TIFR, Mumbai, India.
• [May 2025] Gave a tutorial on Information theory and Differential privacy at TIFR CAM, Bangalore, India.
• [May 2025] Presented my work on Gaussian certified unlearning at IISc, Bangalore, India.
• [Jan 2025] Presented my work on the correlation matrix of spin glasses at IISc and TIFR CAM, Bangalore, India.
• [Oct 2024] Presented my work on hypergraph SBM at Michigan State University, USA.
• [Feb 2024] Our joint work on hypergraph SBM was presented by Yuzhou Gu at the Institute of Advanced Study, USA.

• ORF 445: High Frequency Markets (Spring 2023)
• ORF 387: Networks (Fall 2022)
• MAT 378: Theory of Games (Spring 2023)
• ORF 309: Probability and Stochastic Systems (Fall 2025)
• ORF 245: Fundamentals of Statistics (Fall 2024, 2023)
• ECE 201: Information and Signals (Spring 2026, 2025, 2024)

• Reviewed for Information and Inference (IMAIAI) and Foundations of Computational Mathematics (FOCM).
• Reviewed for IEEE Transactions on Information Theory, NeurIPS 2026, and ICLR 2027.
• Taught a 24-hour course to Master's students, covering probability, martingales and Markov chains.
• Organized a student reading group in High-dimensional Probability and Statistics at Princeton.

• ORF 550: Topics in Probability: Probability in High Dimensions (Fall 2021, Grade A)
• ORF 570: Topics in Statistics and Operations Research: Statistical Machine Learning (Fall 2023, Grade A)
• ECO 519: Advanced Econometrics: Nonlinear Models (Fall 2023, Grade A)
• PHY 535: Phase Transitions and Renormalization Group (Fall 2023, Audited)
• MAT 522: Introduction to Partial Differential Equations (Fall 2022, Audited)
• MAT 572: Topics in Combinatorial Optimization (Fall 2022, Audited)
• MAT 577: Topics in Combinatorics: The Probabilistic Method (Spring 2022, Grade A)
• MAT 586: Computational methods in Cryo-Electron Microscopy (Spring 2024, Audited)
• MAT 589: Topics in Probability, Statistics and Dynamics: Modern Discrete Probability (Fall 2021, Audited)
• MAT 595: Topics in Mathematical Physics (Spring 2024, Audited)
• MAT 597: Introduction to Mathematical Physics (Spring 2023, Audited)

Email: aaradhyapandey@fas.harvard.edu, aaradhyapandeycs@gmail.com

Office: Room 117, Maxwell Dworkin, 33 Oxford Street

Mail: Statistics Department, Harvard University, Massachusetts, 02138, USA