Aritra Mitra

Previously, I was a Postdoctoral Researcher in the Department of Electrical and Systems Engineering, University of Pennsylvania from August 2020 to December 2022, where I worked with Professor George Pappas and Professor Hamed Hassani. I received my Ph.D. in 2020 from the School of Electrical and Computer Engineering, Purdue University, where I was advised by Professor Shreyas Sundaram.

Prior to joining Purdue, I received my M.Tech degree from the Indian Institute of Technology, Kanpur in 2015, and my B.E. degree from Jadavpur University, Kolkata in 2013, both in Electrical Engineering.

I am looking for motivated PhD students to work with me on theoretical problems related to control, optimization, learning, and sequential decision-making under uncertainty (e.g., bandits and reinforcement learning), with a particular focus on multi-agent systems. If you are interested in working with me, please feel free to send me an email.

Research Interests

The broad goal of my research is to enable reliable and efficient learning and decision-making in large-scale distributed systems, while contending with modern challenges related to computation, communication, and adversarial robustness. To meet this goal, my research draws on ideas and tools from Control and Optimization Theory, Statistical Signal Processing, Machine Learning, and Network Science. While my work is theoretically grounded, the theory that I develop is motivated by a variety of application domains: multi-robot systems, wireless sensor networks, federated learning, edge-computing, estimation and control in smart cities and power-grids, and learning in social networks.

My postdoctoral work focused on two main themes: (i) Designing fast and communication-efficient algorithms for the emerging paradigm of Federated Learning; and (ii) Investigating the performance bounds of sequential decision-making problems (e.g., bandits and reinforcement learning) in multi-agent settings. Prior to that, my dissertation made fundamental algorithmic and theoretical contributions to the study of state estimation and statistical inference over networks, subject to worst-case adversarial attacks on certain components. A list of keywords that succinctly describe my past and current research interests is as follows.

• Multi-Agent Reinforcement Learning and Bandits
• Optimization and Statistical Inference
• Federated Learning
• Learning, Control, and Estimation over Networks
• Resilience and Security

Recent Updates

• September 2024: My paper is accepted to the IEEE Transactions on Automatic Control (TAC). This work provides a new, simple, inductive proof technique for a finite-sample analysis of stochastic approximation schemes driven by Markov noise.

• July 2024: 4 papers accepted to the Decision and Control Conference (CDC), 2024. Congrats to my students Sreejeet and Feng on their first CDC papers!

• April 2024: Invited Talk at the Control Seminar, University of Southern California (USC).

• March 2024: Our paper on Compressed Temporal Difference Learning is accepted to the Transactions on Machine Learning Research (TMLR).

• March 2024: Our paper titled Towards Model-Free LQR Control over Rate-Limited Channels has been accepted for an oral presentation (~ 7.5%) at the Learning for Dynamics and Control (L4DC) conference 2024, Oxford!

• March 2024: Invited Talk at the AAAI Spring Symposium on Federated Learning on the Edge, Stanford.

• February 2024: Invited Talk at the Statistics Seminar, Department of Statistics, NC State.

• January 2024: Couple of papers accepted to the American Control Conference (ACC), 2024.

• January 2024: Paper on Stochastic Approximation under Delays accepted to AISTATS 2024.

• January 2024: Paper on Federated On-Policy Reinforcement Learning accepted to ICLR 2024.

• December 2023: Invited (Virtual) Talk in the Network Seminar Series, CNI, Indian Institute of Science (IISc), Bangalore. Video Link

• September 2023: Invited Talk at the ECE Distinguished Speaker Colloquium, NC State.

• June 2023: Our paper on security of networked control systems is accepted to Automatica.

• June 2023: Paper on Federated Reinforcement Learning over Noisy Channels accepted to IEEE Control Systems Letters.

• May 2023: Invited Talk at the SIAM Conference on Optimization (OP23).

• April 2023: Our paper titled Linear Stochastic Bandits over a Bit-Constrained Channel is accepted for an Oral Presentation (~10%) at the Learning for Dynamics and Control Conference (L4DC)!

• March 2023: Invited Talk at the 57th Annual Conference on Information Sciences and Systems (CISS), Johns Hopkins, Baltimore, Maryland.

• February 2023: Organizing a session on Multi-Agent Learning and Decision-Making at ITA 2023, UC San Diego with AbolFazl Hashemi and Arya Mazumdar. Also gave a talk at this session.

• Two new preprints released:
  1. Federated Temporal Difference Learning with Linear Function Approximation Under Environmental Heterogeneity
  2. Temporal Difference Learning with Compressed Updates: Error-Feedback meets Reinforcement Learning
• January 2023: Started as an Assistant Professor at NC State!
• November 2022: Invited talk at the ICON Seminar Series, Purdue University.
• November 2022: Joint talk with Hamed Hassani at Google’s Workshop on Federated Learning and Analytics.
• September 2022: Presented our work on robust collaborative bandits at the Quantifying Uncertainty: Stochastic, Adversarial, and Beyond Workshop at the Simons Institute for the Theory of Computing.

• September 2022: Our paper titled Collaborative Linear Bandits with Adversarial Agents: Near-Optimal Regret Bounds is accepted to NeurIPS 22, New Orleans, USA.
• July 2022: Couple of papers accepted for presentation at the Decision and Control Conference (CDC), Cancun, Mexico.
• March 2022: Invited Talk at the Department of Electrical and Computer Engineering at North Carolina State University.
• March 2022: Invited Talk at the Department of Aerospace Engineering and Engineering Mechanics at the University of Texas, Austin.
• March 2022: Invited Talk at the Department of Electrical and Computer Engineering at the University of Michigan, Ann Arbor.
• February 2022: Invited Talk at the Department of Electrical and Computer Engineering at Georgia Tech.
• February 2022: Invited (Virtual) Talk at the Department of Electrical and Computer Engineering at Rutgers University.
• November 2021: Presented our work at Google’s Workshop on Federated Learning and Analytics. [ Talk ]
• October 2021: Our paper titled Distributed State Estimation Over Time-Varying Graphs: Exploiting the Age-of-Information is accepted to the IEEE Transactions on Automatic Control!
• October 2021: Selected to receive a NeurIPS 2021 Outstanding Reviewer Award given to the top 8% of reviewers who were judged to be instrumental to the review process!
• September 2021: Our paper titled Linear Convergence in Federated Learning: Tackling Client Heterogeneity and Sparse Gradients is accepted to NeurIPS 21!
• September 2021: I gave an invited talk at Google on the Impacts of Statistical Heterogeneity in Federated Supervised Learning and Best-Arm Identification.
• September 2021: Our paper titled On the Computational Complexity of the Secure State-Reconstruction Problem is accepted to Automatica.
• July 2021: Our papers titled Online Federated Learning and Federated Learning with Incrementally Aggregated Gradients are accepted for presentation at the Decision and Control Conference (CDC), 2021, Austin, Texas, USA.
• June 2021: Our paper Distributed Inference with Sparse and Quantized Communication is accepted to the IEEE Transactions on Signal Processing.
• March 2021: Our paper Near-Optimal Data Source Selection for Bayesian Learning is accepted to the 3rd Annual Learning for Dynamics and Control Conference (L4DC) 2021, ETH Zurich, Switzerland.