Fundamental Research

Matsuo Laboratory is developing fundamental technologies related to deep learning, such as deep generative models, deep reinforcement learning, and representation learning. In particular, we have recently been working on world models and their application to robot manipulation as we consider it an important technology to engineer intelligence. Many of our members’ papers have been accepted for publication in top-tier international conferences (ICLR, NeurIPS, ICML), and have received awards, including the Dean’s Award of the Graduate School of Engineering. In a setting environment where researchers with various specialties gather to engage in friendly competition, we provide an environment where motivated students and researchers can play an active role by holding weekly readings of the latest papers, lectures, and exchange of opinions.


Matsuo Laboratory operates TRAIL (Tokyo Robot And Intelligence Lab) in a research and development community aiming to implement intelligence in the real world. In addition to actual robots such as robot arms and mobile manipulators, various simulators, VR-related equipment, 3D printers, and sensors such as depth cameras, motion capture, or tactile sensors are available for use according to your own research and development activities. Tutorials on how to use each hardware and software are provided as needed.

Brain-Inspired Intelligence

Current intelligence technologies based on deep learning have already shown more than comparable performance to humans in areas such as pattern recognition and language processing. However, this is only an extension of a small portion of the human brain’s potential. In reality, current technologies require vast amounts of data and computation. In addition, abilities such as language semantic understanding, self-awareness/metacognition, and hierarchical control have not been implemented at a human-like level. Therefore, the Matsuo Laboratory is engaged in research to unravel the information representation, mechanisms, algorithms, and architecture that support the computational functions of the brain, while taking into account the rapidly accumulating neuroscience findings. Part of our motivation for this research is to reexamine the question of “What is a human being?”. Meanwhile, we aim to connect our findings in world models and robotics with practical engineering technology.

Research Environment

The Matsuo Laboratory has an HPC team dedicated to facilitating deep learning research through state-of-the-art computing resources. We believe that good infrastructure plays a vital role in research, and we provide cutting-edge technology to students and researchers on a daily basis.

The deep learning research team at Matsuo Lab aims to pave the way for the future of artificial intelligence technology.
Our activities are not limited to research alone; we also focus on education and development.

Please click here for more information on our publications.