Siddharth Mitra

I am a PhD student at Yale University in the Department of Computer Science where I am lucky to be advised by Andre Wibisono.

My research focuses on the design and analysis of algorithms for sampling and optimization.

Prior to this, I was an undergraduate and masters student at the Chennai Mathematical Institute as well as a visiting student at the Indian Institute of Science, Bangalore. During this time, I worked with Aditya Gopalan, Himanshu Tyagi, and KV Subrahmanyam.

I can be reached at siddharth.mitra <at> yale <dot> edu.

My Google Scholar.

Research

Authors are listed in alphabetical order, unless denoted by (*).

Characterizing Dependence of Samples along the Langevin Dynamics and Algorithms via Contraction of Φ-Mutual Information
Jiaming Liang, Siddharth Mitra, and Andre Wibisono
IEEE Transactions on Information Theory, to appear.

On the Mixing Time of Unadjusted Hamiltonian Monte Carlo in KL Divergence and Rényi Divergence via One-shot Couplings
Nawaf Bou-Rabee, Siddharth Mitra, and Andre Wibisono
Preliminary version at DynaFront 2025.

On the Convergence of Min-Max Langevin Dynamics and Algorithm [arXiv] [COLT 2025]
Yang Cai, Siddharth Mitra, Xiuyuan Wang, and Andre Wibisono
Conference on Learning Theory (COLT), 2025.

Characterizing Dependence of Samples along the Langevin Dynamics and Algorithms via Contraction of Φ-Mutual Information [arXiv] [COLT 2025 (Extended Abstract)]
Jiaming Liang, Siddharth Mitra, and Andre Wibisono
Conference on Learning Theory (COLT), 2025.

Fast Convergence of Φ-Divergence Along the Unadjusted Langevin Algorithm and Proximal Sampler [arXiv] [ALT 2025]
Siddharth Mitra and Andre Wibisono
International Conference on Algorithmic Learning Theory (ALT), 2025.

Langevin Thompson Sampling with Logarithmic Communication: Bandits and Reinforcement Learning [arXiv] [ICML 2023]
Amin Karbasi, Nikki Kuang, Yian Ma, and Siddharth Mitra
International Conference on Machine Learning (ICML), 2023.

Submodular + Concave [arXiv] [NeurIPS 2021]
Siddharth Mitra (*), Moran Feldman, and Amin Karbasi
Neural Information Processing Systems (NeurIPS), 2021.

On Adaptivity in Information-constrained Online Learning [arXiv] [AAAI 2020]
Siddharth Mitra (*) and Aditya Gopalan
AAAI Conference on Artificial Intelligence (AAAI), 2020.
Preliminary version at OPT 2019.