RIKEN Center for Advanced Intelligence Project Real-World Intelligence Team
Team Director: Tatsuya Harada (D.Eng.)
Research Summary
Our goal is to develop intelligent systems that can integrate various types of information from the real world, such as vision, language, touch, force and action, in order to understand their surroundings and situations, predict future states and act adaptively. The goal is achieved by leveraging a broad range of computer science disciplines. These include machine learning, computer vision, multimodal information processing, robotics and mathematical foundations. The aim is to advance research on "Real-World Intelligence". This is intelligence capable of understanding, predicting and adaptively acting in the real world.
Main Research Fields
- Computer Science
Related Research Fields
- Engineering
Keywords
- Visual Recognition
- Multimodal Information Processing
- Physical AI
Selected Publications
- 1.
Shuhong Liu, Xining Ge, Ziying Gu, Quanfeng Xu, Ziteng Cui, Lin Gu, Xuangeng Chu, Jun Liu, Dong Li, Tatsuya Harada.
"Denoising the Deep Sky: Physics-Based CCD Noise Formation for Astronomical Imaging."
The 19th European Conference on Computer Vision (ECCV), 2026 - 2.
Akio Hayakawa, Yusuke Mukuta, Tatsuya Harada.
"Linear Fusion MultiDiffusion for Fast Training-Free Spherical Panorama Generation"
The 19th European Conference on Computer Vision (ECCV), 2026 - 3.
Dexuan Zhang, Qianyu Zhou, Thomas Westfechtel, Yusuke Mukuta, Tatsuya Harada.
"Bridging Theory and Practice in Source-Free Domain Adaptation via Adversarial Proxy Perturbation"
The 19th European Conference on Computer Vision (ECCV), 2026 - 4.
Ryo Umagami, Tomohiro Hashimoto, Xuangeng Chu, Ziteng Cui, Yusuke Mukuta, Jia Deng, Tatsuya Harada.
"TriSplat: Adaptive Triplane for Sparse-View Large-Scale Scene Reconstruction"
The 19th European Conference on Computer Vision (ECCV), 2026 - 5.
Kazuki Ota, Takayuki Osa, Motoki Omura, Tatsuya Harada.
"Revisiting Regularized Policy Optimization for Stable and Efficient Reinforcement Learning in Two-Player Games."
Forty-third International Conference on Machine Learning (ICML), 2026 - 6.
Haruki Abe, Takayuki Osa, Yusuke Mukuta, Tatsuya Harada.
"Cross-Embodiment Offline Reinforcement Learning for Heterogeneous Robot Datasets"
The Fourteenth International Conference on Learning Representations (ICLR), 2026 - 7.
Naoki Morihira, Amal Nahar, Kartik Bharadwaj, Yasuhiro Kato, Akinobu Hayashi, Tatsuya Harada.
"R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation"
The Fourteenth International Conference on Learning Representations (ICLR), 2026 - 8.
Naoki Shitanda, Motoki Omura, Tatsuya Harada, Takayuki Osa.
"Rethinking Policy Diversity in Ensemble Policy Gradient in Large-Scale Reinforcement Learning"
The Fourteenth International Conference on Learning Representations (ICLR), 2026 - 9.
Ryo Umagami, Liu Yue, Xuangeng Chu, Ryuto Fukushima, Tetsuya Narita, Yusuke Mukuta, Tomoyuki Takahata, Jianfei Yang, Tatsuya Harada.
"Intend to Move: A Dataset for Intention and Scene Aware Human Motion Prediction"
The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025 - 10.
Shuhong Liu, Lin Gu, Ziteng Cui, Xuangeng Chu, Tatsuya Harada.
"I2-NeRF: Learning Neural Radiance Fields Under Physically-Grounded Media Interactions"
The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025
Recent Research Results
Related Links
Lab Members
Principal investigator
- Tatsuya Harada
- Team Director
Core members
- Kenichiro Ando
- Postdoctoral Researcher
- Yasushi Koyama
- Visiting Scientist
- Yusuke Kurose
- Visiting Scientist
- Masatoshi Hidaka
- Visiting Scientist
- Lin Gu
- Visiting Scientist
- Ryosuke Sasaki
- Research Part-time Worker I
- Takehiro Hara
- Research Part-time Worker I
- Ningkang Chang
- Research Part-time Worker I
Contact Information
Engineering Building #2,
7-3-1, Hongo, Bunkyo-ku, Tokyo,
113-8656, Japan
Email: tatsuya.harada@riken.jp