Stefano Ferraro

Papers

Total Citations

H-Index

About

Stefano Ferraro is an emerging researcher working at the intersection of robotics, deep learning, and cognitive-inspired artificial intelligence. His work focuses on two interconnected challenges that are central to modern robotics: computational efficiency in learning systems and intelligent scene understanding. In his 2022 paper, "Computational Optimization of Image-Based Reinforcement Learning for Robotics," Ferraro tackled a critical bottleneck in deploying deep learning models on real robotic hardware — the mismatch between the computational demands of large pretrained models and the limited resources available on physical robotic systems. This contribution, which has garnered 6 citations, offers practical pathways for making image-based reinforcement learning more accessible in real-world robotics applications. His more recent work, "Object-Centric Scene Representations Using Active Inference" (2024, 5 citations), demonstrates a shift toward biologically inspired frameworks, proposing a generative model that allows robots to parse complex environments into meaningful object-level representations — a foundational capability for robot-environment interaction. Though early in his career, Ferraro's research reflects a thoughtful integration of computational pragmatism and cognitive science principles, positioning him as a promising voice in the next generation of robotics and AI researchers.