Papers
2
Total Citations
12
H-Index
2
About
Nicolas Bougie is a researcher advancing the frontiers of reinforcement learning (RL), with a focus on making intelligent agents more autonomous, efficient, and applicable to real-world challenges. His work centers on two key areas: hierarchical learning from sparse rewards and decentralized coordination for large-scale control. In his highly cited 2021 paper, "Hierarchical learning from human preferences and curiosity" (8 citations), Bougie tackles the critical problem of sparse extrinsic rewards in deep RL by integrating human preferences with intrinsic curiosity. This approach enables agents to explore and learn complex behaviors without dense reward signals, a breakthrough for real-world tasks where rewards are naturally scarce. In his 2022 work, "Local Control is All You Need" (4 citations), he addresses the limitations of conventional controllers in process industries by proposing a decentralized, coordinating RL framework. This method eliminates the need for constant parameter tuning and scales to large systems, offering a flexible and generalizable alternative. Bougie’s contributions are shaping the next generation of RL systems that are both human-aware and operationally robust, with his work cited for its practical impact on autonomous decision-making in industrial and interactive settings.
Research Focus
Key Achievements
Top Papers
- 1Hierarchical learning from human preferences and curiosity8 citations · 2021
- 2