About me
Hi! I’m Muni, a Fellow in Artificial Intelligence at Université PSL (Paris Sciences & Letters). I am affiliated with the LAMSADE research center at Université Paris Dauphine-PSL. I work closely with the MILES team members at LAMSADE. My current research interests are in trustworthy machine learning — specifically, robustness, privacy and fairness in machine learning.
I did my PhD in ECE at UW-Madison, advised by Prof. Varun Jog. My thesis was on the theoretical foundations of adversarial robustness in machine learning. Specifically, I explored the surprising connnections between adversarial robustness and optimal transport theory.
Previously, I have worked with Prof. Ambedkar Dukkipati in the Statistics and Machine Learning Group at IISc Bangalore. Before that, I was a software engineer at Samsung R&D Institute in Bangalore. I did my undergrad in Electrical Engineering at IIT Madras.
You can find my CV here.
Papers
- Ilana Sebag, Muni Sreenivas Pydi, Jean-Yves Franceschi, Alain Rakotomamonjy, Mike Gartrell, Jamal Atif, Alexandre Allauzen. “Differentially Private Gradient Flow based on the Sliced Wasserstein Distance for Non-Parametric Generative Modeling” (under submission) 2024.
- Alexandre Vérine, Muni Sreenivas Pydi, Benjamin Negrevergne, Yann Chevaleyre. “Optimal Budgeted Rejection Sampling for Generative Models.” AISTATS 2024.
- Muni Sreenivas Pydi and Varun Jog. “The Many Faces of Adversarial Risk: An Expanded Study.” IEEE Transactions on Information Theory, 2023.
- Alexandre Vérine, Benjamin Negrevergne, Muni Sreenivas Pydi and Yann Chevaleyre. “Precision-Recall Divergence Optimization for Generative Modeling with GANs and Normalizing Flows.” Neurips, 2023.
- Eirini Ioannou, Muni Sreenivas Pydi and Po-Ling Loh, “Robust empirical risk minimization via Newton’s method”, Econometrics and Statistics, 2023.
- Lucas Gnecco-Heredia, Muni Sreenivas Pydi, Yann Chevaleyre, Benjamin Negrevergne and Laurent Meunier. “On the Role of Randomization in Adversarially Robust Classification.” Neurips, 2023.
- Muni Sreenivas Pydi and Varun Jog. “The Many Faces of Adversarial Risk.” Conference on Neural Information Processing Systems (NeurIPS), 2021.
- Muni Sreenivas Pydi and Varun Jog. “Adversarial Risk via Optimal Transport and Optimal Couplings.” IEEE Transactions on Information Theory, 2021.
- Muni Sreenivas Pydi and Varun Jog. “Adversarial Risk via Optimal Transport and Optimal Couplings.” International Conference on Machine Learning (ICML), 2020.
- Muni Sreenivas Pydi, and Vishnu Suresh Lokhande. “Active Learning with Importance Sampling.” NeurIPS Workshop on ML with Guarantees, 2019.
- Muni Sreenivas Pydi, Varun Jog, and Po-Ling Loh. “Graph-Based Ascent Algorithms for Function Maximization.” Communication, Control, and Computing (Allerton), 2018 56th Annual Allerton Conference on. IEEE, 2018.
- Muni Sreenivas Pydi, and Ambedkar Dukkipati. “On Consistency of Compressive Spectral Clustering.” 2018 IEEE International Symposium on Information Theory (ISIT). IEEE, 2018.
- Nagasubramanian, Karthik, and Muni Sreenivas Pydi. “Random access retransmission scheme for power limited nodes.” Communications (NCC), 2017 Twenty-third National Conference on. IEEE, 2017.
- Guha, Ashwin, Muni Sreenivas Pydi, Biswajit Paria, and Ambedkar Dukkipati. “Analytic Connectivity in General Hypergraphs.” arXiv preprint arXiv:1701.04548 (2017).