Matteo Papini

Papers

Total Citations

H-Index

About

Matteo Papini is a researcher specializing in reinforcement learning (RL), with a particular focus on safe and efficient policy optimization methods. His work addresses one of the most pressing challenges in applying RL to real-world settings: ensuring that the learning process itself remains safe and reliable, especially in high-stakes domains like robotics and control systems. Papini's most notable contributions center on safe policy gradient algorithms, explored across multiple iterations of his work on "Smoothing Policies and Safe Policy Gradients," which has collectively garnered over 26 citations. These papers tackle the inherent risks of trial-and-error learning, where unsafe exploration during training can have serious consequences. By developing smoothing techniques for policy optimization, Papini provides principled approaches that constrain the learning process without sacrificing performance, bridging the gap between theoretical RL and practical deployment. His research is particularly relevant for researchers and engineers working at the intersection of machine learning and robotics, where safety guarantees are non-negotiable. Papini's contributions reflect a growing recognition within the RL community that algorithmic performance alone is insufficient — robust, safe learning frameworks are essential for the responsible advancement of autonomous systems in the real world.