Matteo Turchetta
Papers
5
Total Citations
296
H-Index
4
About
Matteo Turchetta is a leading researcher at the intersection of reinforcement learning (RL), robotics, and safety-critical control. His work is defined by a central challenge: how can autonomous systems learn effectively without compromising safety? Turchetta’s major contributions address this through two key themes. First, he pioneered methods for sample-efficient learning, most notably with his highly cited work on "Reinforced Imitation" (198 citations), which combines expert demonstrations with RL to enable mapless navigation with dramatically less data. Second, he is a driving force in the field of safe exploration. His foundational paper on "Safe Exploration in Finite Markov Decision Processes with Gaussian Processes" (68 citations) established a rigorous framework for guaranteeing safety during learning, a critical requirement for deploying robots in the real world. This line of research culminates in his recent work, "GoSafeOpt" (2023), which provides scalable, safe global optimization for dynamical systems, and "Safe Guaranteed Exploration for Nonlinear Systems" (2025), which pushes the frontier by ensuring both safety and complete task coverage. Turchetta’s work is essential reading for anyone seeking to build autonomous systems that are not only intelligent, but also trustworthy and reliable.
Research Focus
Key Achievements
Top Papers
- 1
- 2Safe Exploration in Finite Markov Decision Processes with Gaussian Processes68 citations · 2016
- 3
- 4
- 5Safe Guaranteed Exploration for Nonlinear Systems3 citations · 2025