Naoki Shitanda

PhD student at the University of Tokyo, Japan
#ReinforcementLearning, #RobotLearning

[Naoki Shitanda @NaokiShitanda]

🎉 My latest work on large-scale RL has been accepted to #ICLR2026 !🎉

Appropriately controlled policy diversity improves the learning efficiency of ensemble RL in large-scale environments.

🌐Website: https://t.co/55rSgDx8bw

#RobotRL #RobotLearning #isaacgym #IsaacLab

[ACM SIGGRAPH @siggraph]

Emerging Technologies spotlight: “Katakko: Embodiment of Modular Robots through Automatic Motion Mapping” 🤖

Katakko is a modular robot system that lets users build and embody personalized social agents using communication-oriented modules (display, actuator, mobile, sensor). Its optimization-based mapping algorithm maps human movement to the robot and updates as the system is reconfigured.

Check out more from emerging technologies in the SIGGRAPH 2026 schedule: https://t.co/IHAGRCQP52

Image Credit: Sotaro Yokoi, Naoki Shitanda, Kento Matsuo, and Takuji Narumi, The University of Tokyo.

[Ekin Zorer @ekinomicss]

Misaligned robot exhibiting malicious behavior spotted in #ICLR2026