Members

Okinawa Institute of Science and Technology Graduate University (OIST)

Professor, OIST Cognitive Neuro-Robotics Unit

E-mail : tani1216jp at gmail.com

Prof. Jun Tani Profile

I received my BSc and MSc in Computer Science and Applications from the Department of Computer Science, University of Warwick, U.K., in 2009. After I was selected to receive the Hong Kong PhD Fellowship, I completed my PhD studies at the Department of Computing, The Hong Kong Polytechnic University, in 2015. Following graduation, I worked as a Teaching Fellow and Post-doctoral Fellow in the Cognitive Computing Lab, the Hong Kong Polytechnic University. I have worked in the Tani Unit since October 2017 as a Post-doctoral Scholar, and as a Staff Scientist since April 2021. My background is in data mining and machine learning algorithms for GPUs, and my current research focus is in developing recurrent neural networks for real time robot control in dynamic environments.

I received my master's degree in Computer Science from the University of Bielefeld, Germany, in 2012. For my studies, I worked on control- and power-electronics for compliant actuators and joined the Honda Research Institute Europe GmbH (HRI-EU) for studies on semi-supervised learning and system architectures for the combination of online and offline learning. In 2013, I joined the Research Institute for Cognition and Robotics (CoR-Lab) at the University of Bielefeld and conducted research on the control of soft robots. The topic of my Ph.D. project was: Multi-modal Skill Memories for Online Learning of Interactive Robot Movement Generation. For my experiments on a highly compliant child-like humanoid robot, I visited the Asada Laboratory of the Osaka University, Japan, as a special research student. After passing my doctoral exam in September 2018, I joined the Tani Unit in September 2018 as a post-doctoral scholar and will perform research on: Synthetic Neuro-Robotic Modeling on Coupling of Working Memory, Perception and Action Generation under consideration of Consciousness/Unconsciousness My research interests include cognitive (neuro-) robotic systems, (bio- inspired) machine learning concepts, control of compliant actuators and high- level skill learning.

I got a bachelor's degree in Computer Engineering from the National Autonomous University of Mexico. At some point during my bachelor's I found that studying the brain was more interesting than what I was doing, which eventually led me to Pittsburgh. I did a Math PhD at the University of Pittsburgh, and took courses in the Neural Basis of Cognition at Carnegie Mellon. For the next 10 years I did research on various areas of Computational Neuroscience. My intuition is that understanding the brain will involve grounding cognition in motor control. Moreover, building autonomous agents will be the most valuable source of insights in this effort.

I received my Ph.D. degree in Chemical Engineering from the Dept. of Chemical Engineering and Materials Science at Wayne State University (Detroit, MI, USA), where I served as a teaching and graduate research assistant (2018-2023). My core research area has been in cyber-physical systems using process system engineering framework (researchmap.jp/hoyama). In the realm of cyber-physical systems, I am moving towards exploring the interactions between nonlinear dynamic systems, robotics, and neuroscience in the Cognitive Neurorobotics Research Unit at Okinawa Institute of Science and Technology (2023 - present) under the guidance of Professor Jun Tani.

During my studies in mechanical engineering at the University of Exeter, I found myself interested in artificial intelligence. I later pursued and obtained my MSc in Engineering Mathematics at the University of Bristol. In previous years, I have been working as a research staff studying the application of deep learning in healthcare. I join the CNRU in January 2025 as a research technician.
Since studying for my MSc, I have become a firm believer in the importance of embodied approaches for realizing artificial intelligence that learns and develops efficiently like humans and animals. Therefore, I am very grateful for the opportunity to help out with the wonderful works at CNRU while developing my own research experience in this area. Outside of work, I like Japanese culture, photography and bouldering.

My core research interests have evolved from optimal nonlinear control to robust control (Osuka Lab, 2020), bio-inspired reflective control (Ishikawa Lab, 2021), and learning and control (2022–).

After gaining various academic experiences, including conference presentations, publications, and honors such as the Best Research Award of Osaka Univ. and being selected as a Japanese representative of “YouTube Learning” by Google, I left the university and got hired as a research assistant at Hiroshima University (Fujimoto Lab, 2024), where I worked on brain neurocomputational science research in collaboration with RIKEN. Through a series of fortunate connections, Prof. Tani endorsed me as a member of the CNRU at OIST starting in September 2025.

My ultimate goal is to develop a game-changing framework for controlling spacecraft, which will take me to space. 

This journey is still going on, like a rolling stone.

I received a BSc in applied physics from Keio University, Japan. During my bachelor's, I did a project on thermo-electricity from February 2019 to July 2019 in Kamihara lab at Keio University, and also on superconductivity as a research internship student from July 2019 to January 2020 at Quantum Materials Science Unit (Okada Unit), OIST. During the rotation program in my first year at OIST, I decided to switch my field from applied physics and to take a Ph.D. to neuro-robotics. I joined CNRU in September 2021, and currently doing a project on a humanoid robot, Torobo.

I received a BSc in Mathematics from the University of Chicago in 2017, while also studying computer science and computational neuroscience. Afterwards I worked at a startup developing networked drone comm systems for year, before working at the Angelo Lab at Stanford University writing data analysis tools as a computational tech for three years. I started as a PhD student at the CNRU in January 2022.

I’m interested in autonomy and the mechanisms of open-ended behavioral development. My research is focused on creating working memory systems that support the evolution of graph-like neural programs and data-structures for fast and flexible cognition.

I received a Bachelor's of Science in mathematics from University of California Santa Barbara, then a Master's degree in Applied Data Science from Syracuse University. I am passionate about the relationship between machine learning and the philosophy of consciousness. I also love long distance running, cats, classic video games, cats, and cats!

I joined OIST in 2022, after finishing a Computational Neuroscience MSc in the Canadian Center for Behavioral Neuroscience, and a BSc in Biomedical Engineering from University Pompeu Fabra in Barcelona. My research focuses on discovering the laws that govern learning and cognition, both in biological and artificial beings. Currently, I’m exploring Variational Inference and Generative Models as primitives of complex behavior in agents. Outside the lab, I enjoy playing piano, hiking and doing scuba.

I received a Bachelor's in mechanical engineering from Hosei University, Japan. After graduation, I worked as a technical staff at the National Institute of Advanced Industrial Science and Technology (AIST), studying robot navigation and reinforcement learning with long-term memory. This experience led me to a strong interest in the general principles of the brain behind it and cognitive neuro-robotics research.

University of Missouri (PhD, Philosophy, 2006). Inspired by Ron Sun's Duality of Mind and his founding interest in cognitive social science, my dissertation "Conscience: towards the mechanism of morality" articulated an information processing model of cognitive agency integrating hybrid AI, traditional philosophy and cognitive science. Independently, I work mostly in machine ethics and autonomy, self and social-political system resilience. Currently co-chair, Philosophy of AI, 2018 World Congress of Philosophy, area editor for PhilPapers, reviewer for a number of conferences and journals. Going forward with CNRU, the plan is to work more directly on autonomy and especially fully autonomous artificial agency, extending recent work on "formal consciousness" to "formal autonomy" and building from there.

After a Ph.D. in developmental robotics at the Flowers team in Inria Bordeaux, France, with Pierre-Yves Oudeyer, and a postdoc in neuroscience at in the Mnemosyne team at the Neurodegenerative Diseases Institute in Bordeaux with Nicolas Rougier, I am pursuing a postdoc at OIST focused on developmental processes in cognitive architectures. The goal is to better understand what processes are at play in children during their development, and work toward open-ended, lifelong learning in robots. 

I completed my Ph.D. at OIST in 2024 and am currently a visiting researcher, continuing my collaboration with CNRU.
In my doctoral research, I investigated how the sense of agency—the feeling of authorship over one’s actions—changes during social interactions. I developed a computational model of multimodal social interaction based on the free energy principle and evaluated it through human-robot interaction experiments. As a visiting researcher, I am currently working on a project that explores the mechanisms underlying different dynamics of social interaction using simulation-based experiments.

​Torobo-Humanoid (a.k.a. Minkyu2) is our upper-torso humanoid robot from Tokyo Robotics. The first of its kind, Torobo-Humanoid is strong but tries to actively follow external forces. Also, the RealSense camera is always watching you (not really).

Movie

Torobo-Arm is our arm robot from Tokyo Robotics. With its new specially designed gripper, it is very good at object manipulation. Although big and heavy, Torobo-Arm is a sensitive type and doesn't like hot weather.

OP2 from Robotis is the cutest but also most sensitive robot in our lab. Besides working hard on developing its’ social cognitive abilities, it likes dancing and taking orders form its master device Rakuda.

NAO from Aldebaran, France is the oldest robot member in this unit. It has gone through many exciting experiences, including an imitation game with humans and a ball passing game with another NAO. As NAO is getting older, it is about to retire and give its throne to its successor - OP2.

The NAO6 robot is a humanoid robot designed for research and education. It stands at 58 cm tall and weighs 5.2 kg. It comes equipped with advanced sensors and actuators, including cameras, microphones, touch sensors, and motors, which enable it to perceive and interact with its environment. The robot is powered by sophisticated software that allows for natural and intuitive interaction with humans, making it an ideal platform for studying human-robot interaction, artificial intelligence, and robotics programming.

Dimitri is an open-software & open-hardware robot torso equipped with modular low cost compliant joints designed by Rodrigo Da Silva. Its elastic actuators and hard materials make it a tough robot with flexible joints. Unfortunately, Dimitri is currently sick and needs a passionate caregiver to heal it.