Group Members
Current Graduate Students
Yashwanthi Anand (PhD in CS)
Pulkit Rustagi (PhD in ROB)
Jeff Jewett (PhD in CS)
William Sollow (PhD in AI)
Alumni
Undergraduate students
Malhar Milind Damle (Honor’s thesis: Safe Navigation in Highway Environment Using Behavioral Cloning )
Kunal Chopra
Avi Desi
Selected Projects
Large-scale data-driven decision-making
Autonomous systems and intelligent decision-support systems are being deployed in many real-world settings. Harnessing the full benefits of these systems requires addressing various technological challenges. The goal is to design systems that are reliable, can reason under complex real-world settings, and improve the well-being of the community. We use a combination of AI planning, optimization techniques, and game-theoretic approaches to achieve this. Our emergency response system and traffic patrolling systems have been deployed in Singapore. Media Coverage.
Relevant papers: IAAI 2014, AAAI 2015, SPARK 2017, ITSC 2019.
Safe and Reliable Autonomous Systems
Autonomous agents acting in the real-world often operate based on imprecise models. The limited fidelity of the model used for reasoning creates negative side effects and other safety concerns that may not be easy to identify at design time. The goal is to design autonomous systems that can adapt and learn to mitigate undesirable impacts when deployed.
Relevant papers: IJCAI 2020, AAMAS 2021, CHI 2021, AI Magazine 2022, JAIR 2022, ICRA 2022, AAAI 2023, IJCAI 2023.
3. Fair and interpretable decision-support systems
Decision support for real-world problems often employs complex reasoning models to improve the accuracy of decisions. As these systems are increasingly used in high-stake domains, it is critical to ensure that these systems produce fair and interpretable results. Due to incomplete specification, the results may be biased. When the results are not interpretable, it is difficult for the user to quickly identify the biases and know when to trust the system. How to design accurate, fair, and interpretable reasoning models for decision support?