Davide Corsi
Papers
12
Total Citations
131
H-Index
6
About
Davide Corsi is a researcher at the forefront of safe artificial intelligence, specializing in deep reinforcement learning (DRL), formal verification, and autonomous robotic systems. His work addresses one of the most pressing challenges in modern AI: ensuring that learning-based systems behave reliably and safely when deployed in high-stakes, real-world environments. Corsi's most influential contribution — his 2021 paper on safe reinforcement learning for autonomous robotic-assisted surgery (38 citations) — demonstrates how DRL can automate delicate surgical subtasks while maintaining rigorous safety guarantees. This work exemplifies his broader research agenda: applying formal verification techniques to neural networks operating in safety-critical domains, from medical robotics to aquatic navigation and autonomous exploration. His 2023 paper on verifying learning-based robotic navigation systems (18 citations) further extends this vision, tackling the known vulnerability of DRL policies to unexpected bugs. Beyond verification, Corsi has made notable contributions to curriculum learning, constrained reinforcement learning, and hybrid evolutionary-DRL approaches, reflecting a versatile and methodologically rich research profile. With over 120 cumulative citations across his published work and applications spanning surgical robotics, colonoscopy navigation, and industrial manipulators, Corsi is emerging as a significant voice in the responsible deployment of autonomous intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Verifying Learning-Based Robotic Navigation Systems18 citations · 2023
- 3Benchmarking Safe Deep Reinforcement Learning in Aquatic Navigation14 citations · 2021
- 4Curriculum learning for safe mapless navigation13 citations · 2022
- 5
- 6Genetic Soft Updates for Policy Evolution in Deep Reinforcement Learning12 citations · 2021
- 7
- 8
- 9Verifying Learning-Based Robotic Navigation Systems4 citations · 2022
- 10