Shalini Ghosh
Papers
1
Total Citations
6
H-Index
1
About
Shalini Ghosh is a leading researcher in trusted machine learning, with a focus on ensuring the safety and reliability of AI systems deployed in high-stakes environments. Her work centers on the intersection of formal methods, reinforcement learning, and cyber-physical systems, particularly for applications like autonomous driving, surgical robotics, and cybersecurity. In her highly cited 2018 paper, "Model, Data and Reward Repair: Trusted Machine Learning for Markov Decision Processes," Ghosh introduced a groundbreaking paradigm for providing high-level guarantees—such as safety and liveness—in ML-driven decision-making. This work has been foundational in the emerging field of trustworthy AI, earning her recognition for bridging the gap between rigorous formal verification and practical machine learning. With over 6 citations on this seminal paper alone, Ghosh’s contributions are shaping how researchers and engineers build AI systems that are not only powerful but also provably safe. Her innovative approach to "repairing" models, data, and reward structures has made her a key voice in the push for accountable, robust AI—a critical need as these technologies become increasingly embedded in our daily lives.
Research Focus
Key Achievements
Top Papers
- 1