Max Robotics — Global Service Robot Directory + US Certification

About

Yarden As is a researcher working at the intersection of safe reinforcement learning, probabilistic machine learning, and real-world robotic applications. Their work addresses one of the most pressing challenges in deploying autonomous systems: ensuring safety under uncertainty, whether in high-stakes medical environments or complex control tasks with unknown constraints. Perhaps their most impactful contribution is SafeRPlan, a safe deep reinforcement learning framework for intraoperative planning of pedicle screw placement during spinal fusion surgery — a domain where millimeter-level precision can mean the difference between success and catastrophic harm. This work, garnering 16 citations since 2024, demonstrates how safe RL can be meaningfully applied to surgical robotics. Complementing this, their earlier work on log barriers for safe black-box optimization offers principled methods for learning safety constraints from noisy, real-world feedback, with direct applications in manufacturing and robotics. As also contributes to meta-reinforcement learning through PACOH-RL, enabling data-efficient adaptation to changing system dynamics — a critical capability for deployable robotic systems. Across their portfolio, a clear theme emerges: bridging rigorous theoretical safety guarantees with practical, high-consequence applications, making their research increasingly relevant as autonomous systems enter clinical and industrial settings.