Michael Teti
Papers
1
Total Citations
5
H-Index
1
About
Michael Teti’s research lies at the intersection of autonomous systems, behavioral cloning, and deep neural network interpretability. In his most-cited work, a 2021 study on behaviorally-cloned deep neural networks in autonomous steering, Teti conducted a controlled investigation into how these models replicate human driving decisions. This work, garnering 5 citations, is notable for its rigorous experimental design—a rarity in the early, often anecdotal, literature on imitation learning for self-driving cars. Teti’s contribution is a methodological blueprint: by systematically isolating network behaviors under varied steering scenarios, he revealed critical failure modes and biases in cloned policies, advancing the safe deployment of AI in real-world navigation. His findings have informed subsequent research on adversarial robustness and model validation in autonomous systems. Though early in his career, Teti’s focus on transparent, reproducible evaluation marks him as a careful voice in a field often driven by performance metrics alone. For students and researchers, his work underscores the importance of controlled testing in understanding—and trusting—learned autonomous behaviors.
Research Focus
Key Achievements
Top Papers
- 1