I am a third-year PhD student working under the guidance of
Professor Shlomo Zilberstein at
UMass Amherst.
My doctoral research is centered around enhancing the safety, compliance, and explainability of sequential decision-making systems. Over the course of my academic journey,
I have developed a diverse skill set in relevant areas such as Reinforcement Learning (RL), Learning from Demonstrations (LfD), Explanation Generation,
Shared Autonomy, and Continual Learning. Currently, my focus lies on two intriguing topics: learning discrete representations such as Automata,
Binary Decision Diagrams, and Tree structures for transparent and incremental learning, and developing sequential perturbation-based explanations for RL agents.
I am actively seeking a research internship opportunity in the aforementioned domains for the summer of 2024. For more detailed information about my qualifications and experiences, you can access my CV here: [CV].
News:
- December 2023: A full paper accepted in AAMAS 2023.
- December 2023: Passed PhD portfolio with distinction and became a PhD candidate.
- July 2023: A full paper accepted in IROS 2023.
- April 2023: A full paper accepted in IJCAI 2023.
- January 2023: A full paper and an abstract accepted in AAMAS 2023.
PhD Research:
- Saaduddin Mahmud, Marcell VazquezChanlatte, Stefan Witwicki and Shlomo Zilberstein.
Explaining the Behavior of POMDP-based Agents Through the Impact of Counterfactual Information. (AAMAS 2024).
- Connor Basich*,Saaduddin Mahmud*, and Shlomo Zilberstein.
Learning Constraints on Autonomous Behavior from Proactive Feedback (IROS 2023).
- Saaduddin Mahmud, Sandhya Saisubramanian, and Shlomo Zilberstein.
Explanation-Guided Reward Alignment.(IJCAI 2023). [IJCAI]
- Saaduddin Mahmud, Samer B. Nashed, Claudia V. Goldman, and Shlomo Zilberstein.
Estimating Causal Responsibility for Explaining Autonomous Behavior(EXTRAAMAS-2023 @ AAMAS-2023). [SPRINGER]
- Saaduddin Mahmud*, Connor Basich*, and Shlomo Zilberstein.
Semi-Autonomous Systems with Contextual Competence Awareness(AAMAS 2023). [ACM]
- Saaduddin Mahmud, Sandhya Saisubramanian, and Shlomo Zilberstein.
REVEALE: A Framework for Reward Verification and Learning.(SafeAI @ AAAI-23, Best Paper Award Nomination (Top 4)). [CEUR-WS]
- Samer B. Nashed, Saaduddin Mahmud, Claudia V. Goldman, and Shlomo Zilberstein.
Causal Explanations for Sequential Decision Making Under Uncertainty(AAMAS 2023 (Ext. Abs.) and Under Review for JAIR).
[ARXIV]
Undergrad Research:
- Saaduddin Mahmud,
Md. Mosaddek Khan,
Moumita Choudhury,
Long Tran-Thanh,
and Nicholas R. Jennings.
Learning Optimal Temperature Region for Solving Mixed Integer Functional DCOPs. In the proceedings of the 29th
International Joint Conference on Artificial Intelligence (IJCAI 2020).
[ARXIV]
[IJCAI]
- Saaduddin Mahmud,
Moumita Choudhury,
Md. Mosaddek Khan,
Long Tran-Thanh,
and Nicholas R. Jennings.
AED: An Anytime Evolutionary DCOP Algorithm. In the proceedings of the 19th International Conference on Autonomous
Agents and Multi-Agent Systems (AAMAS 2020).
[ARXIV]
[IFAAMAS]
[ACM]
- Moumita Choudhury, Saaduddin Mahmud, and
Md. Mosaddek Khan. A Particle Swarm Based Algorithm for Functional Distributed Constraint Optimization Problems.
In the proceedings of the Thirty-Fourth AAAI Conference on
Artificial Intelligence (AAAI 2020).
[AAAI]
- Saaduddin Mahmud and Moumita Choudhury (Equal Contribution). Applying Population-Based Algorithms to Solve Large (F)DCOPs.
Undergraduate Thesis, Department of Computer Science and Engineering, University Of Dhaka, 2020.
- Rafid Amir Mahmud, Fahim Faisal, Saaduddin Mahmud,
and Md. Mosaddek Khan.
A Simulation-Based Online Planning Algorithm for Multi-Agent Cooperative Environments.
[AAMAS 2022 (Ext. Abs.)]
[ARXIV]
Softwares:
- AL.GO: A project written in Java to visualize well-known algorithms and help students learn them faster.
[Youtube, 2017]
- Step-by-step algorithm visualizer.
- Contains codes, and problem links on specific topics to help students learn faster.
- MuSyc: An android application for music synchronization across mobile devices.
[Github, 2017]
- Music synchronization across different mobiles.
- Social-network for sharing music.
- EasyML: A web application written in Python for automated data visualization and classification.
[Github, 2018]
- Automated data visualization and classification.
- Fast hyper-parameter optimization for different classifiers.
Blogs: