Francois Belletti
Papers
2
Total Citations
157
H-Index
2
About
Francois Belletti is a researcher at the intersection of machine learning and intelligent transportation systems, with a particular focus on applying deep reinforcement learning to real-world cyberphysical engineering challenges. His most recognized contribution, "Expert Level Control of Ramp Metering Based on Multi-Task Deep Reinforcement Learning" (2017), has accumulated over 155 citations and represents a landmark achievement in autonomous traffic management. In this work, Belletti demonstrated that the same reinforcement learning breakthroughs powering robots in arcade games and physical manipulation tasks could be effectively harnessed to optimize highway ramp metering — a complex, high-stakes control problem traditionally requiring extensive human expertise. By framing ramp metering as a multi-task deep RL problem, he helped establish a new paradigm for data-driven traffic control that moves beyond hand-crafted rules toward adaptive, learned policies. His research sits at a compelling crossroads of artificial intelligence and infrastructure engineering, offering practical pathways for smarter, more responsive transportation networks. Belletti's work has influenced both the academic RL community and practitioners developing next-generation intelligent transportation systems, making him a notable contributor to applied machine learning in urban and highway mobility contexts.
Research Focus
Key Achievements
Top Papers
- 1
- 2