Max Robotics — Global Service Robot Directory + US Certification

About

Kristian Hartikainen is a leading researcher at the intersection of deep reinforcement learning (RL) and real-world robotics. His work is best known for advancing sample-efficient, robust RL algorithms and demonstrating their application on complex, low-cost robotic platforms. Hartikainen is a co-author of the highly influential **Soft Actor-Critic (SAC)**, a model-free deep RL algorithm that has become a cornerstone of the field, amassing nearly 2,000 citations for its ability to overcome high sample complexity and hyperparameter brittleness. He further extended this work in the widely-used "SAC Applications" paper. Beyond algorithmic theory, Hartikainen is a driving force in bringing RL to physical systems. He led the development of **ROBEL**, an open-source platform of low-cost robots designed to accelerate real-world RL research. His most recent landmark achievement is teaching a low-cost, miniature bipedal robot to play agile, one-versus-one soccer using deep RL—a feat that required synthesizing sophisticated, safe movement skills in dynamic environments. This work, published in 2024, has already garnered significant attention. Through his contributions to both foundational algorithms and practical robotic systems, Hartikainen is helping to bridge the gap between simulated RL and robust, real-world deployment.