Deepak-George Thomas
Papers
1
Total Citations
4
H-Index
1
About
Deepak-George Thomas is a rising researcher at the intersection of software engineering and artificial intelligence, with a primary focus on the testing and reliability of deep reinforcement learning (RL) systems. His most notable contribution is the development of **μPRL**, a mutation testing pipeline specifically designed for deep RL agents, which was published in 2025 and has already garnered early citations. This work addresses a critical gap: as RL is increasingly deployed in high-stakes domains like autonomous driving and robotics, traditional testing methods fall short. μPRL introduces a systematic way to inject real-world faults into RL models, enabling researchers to evaluate the robustness of their agents before deployment. By grounding his approach in actual fault patterns rather than synthetic errors, Thomas provides a more realistic assessment of model resilience. His research is particularly impactful for practitioners building safety-critical autonomous systems, offering a practical tool to measure test adequacy. With his work already cited in the emerging literature on trustworthy AI, Deepak-George Thomas is establishing himself as a key voice in ensuring that RL systems are not only powerful but also dependable in real-world environments.
Research Focus
Key Achievements
Top Papers
- 1