Max Robotics — Global Service Robot Directory + US Certification

About

Nevena Lazic is a leading researcher at the intersection of reinforcement learning and high-speed robotics, best known for her groundbreaking work in robotic table tennis. Her research focuses on developing sample-efficient, model-free algorithms that enable robots to perform complex, dynamic tasks in real-world environments. Lazic’s major contributions include pioneering the use of evolutionary search methods with CNN-based policies to control robot joints at 100Hz, allowing robots to return table tennis balls with remarkable precision. Her 2020 paper on this work has garnered 35 citations, while her comprehensive 2023 case study on the same system has been cited 17 times. Notably, Lazic achieved a historic milestone in 2025 with the first learned robot agent to reach amateur human-level performance in competitive table tennis, a feat documented in her most recent highly cited work. Beyond robotics, she has advanced the theoretical foundations of reinforcement learning, with her 2020 paper on provably efficient adaptive approximate policy iteration earning 9 citations. Lazic’s hierarchical policy design for self-play learning, published in 2018, further demonstrates her commitment to sample-efficient training methods. Her work not only pushes the boundaries of what robots can achieve in high-speed, interactive settings but also provides a blueprint for bridging theory and practice in reinforcement learning.