Home

About Me

I have been an Assistant Professor at the Department of Electrical and Computer Engineering at New Jersey Institute of Technology (NJIT) since Fall 2023..

Research Interests: My overarching goal is to develop safe intelligent decision-making. While Reinforcement Learning (RL) is increasingly used in many applications, my current research focuses on safety aspects. In particular, my current research seeks to develop an implementable safe architecture for RL. The challenges are plenty: (i) how to develop a safe exploration framework, (ii) how to ensure sim-to-real gap is closed, (iii) how to adapt to changing dynamic environment while being safe, (iv) how to learn safe and yet reward maximizing policies from offline data. Please see my recent publication, and research projects

In general, I am interested in mathematical understanding of deep learning/learning models. 

Note to Students: I will be looking for motivated, hard-working Ph.D. students in the area of Applied Mathematics, Machine Learning, Wireless Communication, Control Theory, and Computer Networks. Please apply to the Ph.D. program and send your CVs at arnob.ghosh@njit.edu. If you are an undergrad at the NJIT please drop a mail with your list of potential interests, and the coursework you have done.

Past Experience: Before joining NJIT, I was a Research Scientist at the Electrical and Computer Engineering Department of The Ohio State University. I was also part of the NSF AI-EDGE Institute (https://aiedge.osu.edu/).  I was hosted by Ness Shroff since June 2021.  Prior to that, I was an Assistant Professor at the Industrial Engineering and Operations Research (IEOR) Program of Mechanical  Engineering Department at Indian Institute of Technology-Delhi (IIT-Delhi) since August 2019.  At IIT-Delhi, I was also associated with the School of Artificial Intelligence. While serving as an Assistant Professor, I also spent at the Imperial College as a Research Associate where I was hosted by Thomas Parisini. Prior to joining IIT-Delhi, I was a Post-doctoral research associate in the School of Industrial Engineering at Purdue University from August 2016 to July 2019. There, I was hosted by Dr. Vaneet Aggarwal. 

I also spent 3 months (May'2017- August' 2017) at Northwestern University in the Department of Electrical Engineering and Computer Science with Prof. Randall Berry.  I have been privileged to work with Eitan Altman during the summer of 2014 at INRIA, Sophia Antipolis, France. 

I completed my Ph.D. in August, 2016 from the Electrical and System Engineering Department, University of Pennsylvania where I was advised by Prof. Saswati Sarkar. Before joining UPENN, I have studied Electronics & Telecommunication Engineering at Jadavpur University from 2007-2011.  

 My CV is here. 

 My Google Scholar page is here.

News (10 Latest ones)

1. My Ph.D. student Sourav Ganguly has his first Ph.D. paper accepted at the IEEE Conference on Decision and Control (CDC), Congrats!! Details will follow soon..

2. 1 paper on non-stationary RL-based mobile beamforming has been accepted at ACM MobiHoc. Please see the demo https://www.dropbox.com/scl/fi/ogv10le30cck1goxh3voa/ris_bandit_demo.mp4?rlkey=nd8orz1sktmnvdk58udbd5b6v&e=2&dl=0

3. 2 papers accepted at ICML'2025. (i) We provide the first regret and violation bound for the risk-constrained MDP. https://openreview.net/pdf?id=5s7D7FPuTc

(ii) We also provide the first safe exploration guarantee for linear MDP with instantaneous constraints https://arxiv.org/pdf/2502.18655

4. Paper on `Communication Efficient Asynchronous Stochastic Gradient Descent)' has been accepted at the INFOCOM'2025.  

5. Paper on `Safe offline Reinforcement Learning' has been accepted at the NeurIPS'24. 

6. Paper on `Distributionally Robust safe Reinforcement learning' has been presented at the RLSW workshop at RLC'24.

7. Paper on ``Towards Achieving Sub-linear Regret and Hard Constraint Violation in Model-free RL" has been accepted at AISTATS'24.

8.  Paper on ``Achieving fairness in multi-agent Reinforcement Learning" has been accepted at ICLR'24.

9. I am going to teach Reinforcement Learning course in Spring 2024. Please register for the course or contact me if you need any information.

10. Nov. 2023: Selected as co-chair of the Robotics and Automation Chapter of IEEE North Jersey Section.

Education

• Ph.D.  Electrical and Systems Engineering, University Of Pennsylvania, Sept. 2011-August2016.

• M.S., Electrical Engineering, University Of Pennsylvania, May 2013.

• B.E., Electronics and Telecommunication Engineering, Jadavpur University, May 2011.