suzuki masahiro 鈴木雅大

… Alternative Names

Suzuki Masahiro 鈴木雅大

Less

Researcher Number	30823885
Other IDs
Affiliation (Current)	2026: 東京大学, 大学院工学系研究科(工学部), 特任講師
Affiliation (based on the past Project Information) *help	2026: 東京大学, 大学院工学系研究科(工学部), 特任講師 2023 – 2025: 東京大学, 大学院工学系研究科(工学部), 特任助教 2018 – 2019: 東京大学, 大学院工学系研究科(工学部), 特任研究員
Review Section/Research Field	Principal Investigator Basic Section 61030:Intelligent informatics-related / Transformative Research Areas, Section (IV) / 1001:Information science, computer engineering, and related fields Except Principal Investigator Basic Section 62010:Life, health and medical informatics-related
Keywords	Principal Investigator 深層生成モデル / マルチモーダル学習 / 世界モデル / 自由エネルギー原理 / 共有表現学習 / 深層学習 Except Principal Investigator 個別化医療 / AI / 人工知能 / 深層学習 / マルチモーダル

タスク駆動による潜在空間選択機構を持つ世界モデルの構築Principal Investigator
- Principal Investigator
  
  鈴木雅大
- Project Period (FY)
  2026 – 2028
- Research Category
  
  Grant-in-Aid for Scientific Research (C)
- Review Section
  
  Basic Section 61030:Intelligent informatics-related
- Research Institution
  The University of Tokyo
個別化医療の実現を目指したマルチモーダル汎用モデル開発
- Principal Investigator
  
  小寺聡
- Project Period (FY)
  2024 – 2025
- Research Category
  
  Grant-in-Aid for Scientific Research (B)
- Review Section
  
  Basic Section 62010:Life, health and medical informatics-related
- Research Institution
  The University of Tokyo
A Study on Deep Generative Models based on The Free Energy PrinciplePrincipal Investigator
- Principal Investigator
  
  鈴木雅大
- Project Period (FY)
  2023 – 2027
- Research Category
  
  Grant-in-Aid for Transformative Research Areas (A)
- Review Section
  
  Transformative Research Areas, Section (IV)
- Research Institution
  The University of Tokyo
A study on shared representation learning considering the uncertainty of each modalityPrincipal Investigator
- Principal Investigator
  
  Suzuki Masahiro
- Project Period (FY)
  2018 – 2019
- Research Category
  
  Grant-in-Aid for Research Activity Start-up
- Review Section
  
  1001:Information science, computer engineering, and related fields
- Research Institution
  The University of Tokyo

All 2024 2023 2020 2019 2018

All Journal Article Presentation

[Journal Article] GIRL: Reward Function Learning Framework Independent of Text Generator Samples for Reinforcement Learning in Text Generation Tasks2024
- Author(s)
  冨山翔司、鈴木雅大、落合桂一、松尾豊
- Journal Title
  
  電子情報通信学会論文誌D 情報・システム
  
  Volume: J107-D Issue: 5 Pages: 348-358
- DOI
  10.14923/transinfj.2023DEP0009
- ISSN
  1880-4535, 1881-0225
- Year and Date
  2024-05-01
- Language
  Japanese
- Peer Reviewed
- Data Source
  KAKENHI-ORGANIZER-23H04972, KAKENHI-PLANNED-23H04974
[Journal Article] Symbol Emergence and Representation Learning by Integrating GPLVM and Neural Network2024
- Author(s)
  Nakamura Tomoaki、Suzuki Masahiro、Taniguchi Akira、Taniguchi Tadahiro
- Journal Title
  
  Journal of the Robotics Society of Japan
  
  Volume: 42 Issue: 7 Pages: 696-699
- DOI
  10.7210/jrsj.42.696
- ISSN
  0289-1824, 1884-7145
- Language
  Japanese
- Peer Reviewed
- Data Source
  KAKENHI-PLANNED-23H04835, KAKENHI-ORGANIZER-23H04972, KAKENHI-PLANNED-23H04974, KAKENHI-PROJECT-21H04904
[Journal Article] Development and analysis of medical instruction-tuning for Japanese large language models2024
- Author(s)
  Sukeda Issey、Suzuki Masahiro、Sakaji Hiroki、Kodera Satoshi
- Journal Title
  
  Artificial Intelligence in Health
  
  Volume: 1 Issue: 2 Pages: 107-107
- DOI
  10.36922/aih.2695
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K28181
[Journal Article] A Method for Detecting DeepFake Videos Using Angular Information of Faces2023
- Author(s)
  蔭山智、鈴木雅大、落合桂一、松尾豊
- Journal Title
  
  電子情報通信学会論文誌D 情報・システム
  
  Volume: J106-D Issue: 5 Pages: 317-327
- DOI
  10.14923/transinfj.2022DEP0006
- ISSN
  1880-4535, 1881-0225
- Year and Date
  2023-05-01
- Language
  Japanese
- Peer Reviewed
- Data Source
  KAKENHI-PLANNED-23H04974
[Journal Article] Learning global spatial information for multi-view object-centric models2023
- Author(s)
  Kobayashi Yuya、Suzuki Masahiro、Matsuo Yutaka
- Journal Title
  
  Advanced Robotics
  
  Volume: 37 Issue: 13 Pages: 828-839
- DOI
  10.1080/01691864.2023.2183780
- Peer Reviewed
- Data Source
  KAKENHI-PLANNED-23H04974
[Journal Article] World models and predictive coding for cognitive and developmental robotics: frontiers and challenges2023
- Author(s)
  Taniguchi Tadahiro、Murata Shingo、Suzuki Masahiro、Ognibene Dimitri、Lanillos Pablo、Ugur Emre、Jamone Lorenzo、Nakamura Tomoaki、Ciria Alejandra、Lara Bruno、Pezzulo Giovanni
- Journal Title
  
  Advanced Robotics
  
  Volume: 37 Issue: 13 Pages: 780-806
- DOI
  10.1080/01691864.2023.2225232
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-ORGANIZER-23H04972, KAKENHI-PLANNED-23H04974
[Journal Article] 自由エネルギー原理と深層学習—世界モデルを軸として—2023
- Author(s)
  鈴木雅大
- Journal Title
  
  Journal of the Japanese Society for Artificial Intelligence
  
  Volume: 38 Issue: 6 Pages: 796-804
- DOI
  10.11517/jjsai.38.6_796
- ISSN
  2188-2266, 2435-8614
- Year and Date
  2023-11-01
- Language
  Japanese
- Data Source
  KAKENHI-ORGANIZER-23H04972, KAKENHI-PLANNED-23H04974
[Journal Article] Scene Interpretation by Deep Generative Model Utilizing Information of Backgrounds2023
- Author(s)
  Kobayashi Yuya、Suzuki Masahiro、Matsuo Yutaka
- Journal Title
  
  Transactions of the Japanese Society for Artificial Intelligence
  
  Volume: 38 Issue: 3 Pages: E-L35_1-12
- DOI
  10.1527/tjsai.38-3_E-L35
- ISSN
  1346-0714, 1346-8030
- Year and Date
  2023-05-01
- Language
  Japanese
- Peer Reviewed
- Data Source
  KAKENHI-PLANNED-23H04974
[Journal Article] Neuro-SERKET: Development of Integrative Cognitive System through the Composition of Deep Probabilistic Generative Models2020
- Author(s)
  Tadahiro Taniguchi, Tomoaki Nakamura, Masahiro Suzuki, Ryo Kuniyasu, Kaede Hayashi, Akira Taniguchi, Takato Horii, Takayuki Nagai
- Journal Title
  
  New Generation Computing
  
  Volume: 38 Issue: 1 Pages: 23-48
- DOI
  10.1007/s00354-019-00084-w
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-16H06569, KAKENHI-PROJECT-18H03308, KAKENHI-PROJECT-19K21527
[Journal Article] 服の領域を考慮した写真上の人物の自動着せ替えに関する研究2019
- Author(s)
  久保静真，岩澤有祐，鈴木雅大，松尾豊
- Journal Title
  
  情報処理学会論文誌
  
  Volume: 60 Pages: 870-879
- NAID
  170000150210
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-19K21527
[Journal Article] 深層生成モデルを用いた半教師ありマルチモーダル学習2018
- Author(s)
  鈴木雅大，松尾豊
- Journal Title
  
  情報処理学会論文誌
  
  Volume: 59 Pages: 2261-2278
- NAID
  170000149944
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] Object-Centric Temporal Consistency via Conditional Autoregressive Inductive Biases2024
- Author(s)
  Cristian Meo, Akihiro Nakano, Mircea Lic, Aniket Didolkar, Masahiro Suzuki, Anirudh Goyal, Mengmi Zhang, Justin Dauwels, Yutaka Matsuo, Yoshua Bengio
- Organizer
  NeurIPS 2024 Workshop on Compositional Learning: Perspectives, Methods, and Paths Forward
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] The Embodied World Model Based on LLM with Visual Information and Prediction-Oriented Prompts2024
- Author(s)
  Wakana Haijima, Kou Nakakubo, Masahiro Suzuki, Yutaka Matsuo
- Organizer
  The Forty-first International Conference on Machine Learning (ICML2024) Multi-modal Foundation Model meets Embodied AI Workshop
- Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] 動画生成AIと世界モデル2024
- Author(s)
  鈴木雅大
- Organizer
  生成AI革命　言語×視覚×動作の融合がもたらす次なる地平
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] SSM Meets Video Diffusion Models: Efficient Video Generation with Structured State Spaces2024
- Author(s)
  Yuta Oshima, Shohei Taniguchi, Masahiro Suzuki, Yutaka Matsuo
- Organizer
  The Twelfth International Conference on Learning Representations (ICLR2024) 5th Workshop on practical ML for limited/low resource settings
- Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] 物体中心表現を用いた世界モデルの獲得2024
- Author(s)
  中野聡大, 鈴木雅大, 松尾豊
- Organizer
  2024年度人工知能学会全国大会
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] Deep Generative Models and World Models2024
- Author(s)
  Masahiro Suzuki
- Organizer
  1st Digital Brain Workshop
- Invited
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] ADOPT: Modified Adam Can Converge with Any β2 with the Optimal Rate2024
- Author(s)
  Shohei Taniguchi, Keno Harada, Gouki Minegishi, Yuta Oshima, Seong Cheol Jeong, Go Nagahara, Tomoshi Iiyama, Masahiro Suzuki, Yusuke Iwasawa, Yutaka Matsuo
- Organizer
  Advances in Neural Information Processing Systems 37 (NeurIPS 2024)
- Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] Perspectives on World Models and Predictive Codings in Cognitive Robotics2023
- Author(s)
  Masahiro Suzuki
- Organizer
  The 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2023) Workshop
- Invited / Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] 世界モデルベースマルチエージェント強化学習におけるエージェント間の公平性を考慮した経路計画手法の提案2023
- Author(s)
  青木瑞穂, 藤重天真, 塚本慧, 藤本昌也, 鈴木雅大
- Organizer
  2023年度人工知能学会全国大会（第37回）
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] Interaction-Based Disentanglement of Entities for Object-Centric World Models2023
- Author(s)
  Akihiro Nakano, Masahiro Suzuki, Yutaka Matsuo
- Organizer
  The Eleventh International Conference on Learning Representations (ICLR 2023)
- Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] 潜在拡散モデルを用いた世界モデルの提案2023
- Author(s)
  山蔦栄太郎, 内山史也, 関戸麗矢, 川原雄登, 鈴木雅大, 松尾豊
- Organizer
  2023年度人工知能学会全国大会（第37回）
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] End-to-end Training of Deep Boltzmann Machines by Unbiased Contrastive Divergence with Local Mode Initialization2023
- Author(s)
  Shohei Taniguchi, Masahiro Suzuki, Yusuke Iwasawa, Yutaka Matsuo
- Organizer
  The 40th International Conference on Machine Learning (ICML 2023)
- Int'l Joint Research
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] 反復償却推論によるマルチモーダル情報統合の改善2023
- Author(s)
  大島佑太, 鈴木雅大, 松尾豊
- Organizer
  2023年度人工知能学会全国大会（第37回）
- Data Source
  KAKENHI-PLANNED-23H04974
[Presentation] Dual Space Learning with Variational Autoencoders2019
- Author(s)
  Hirono Okamoto, Masahiro Suzuki, Itto Higuchi, Shohei Ohsawa, Yutaka Matsuo
- Organizer
  ICLR workshop
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] Dual Space Learning with Variational Autoencoders2019
- Author(s)
  Hirono Okamoto, Masahiro Suzuki, Itto Higuchi, Shohei Ohsawa, Yutaka Matsuo
- Organizer
  Workshop on Deep Generative Models for Highly Structured Data, International Conference on Learning Representation
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] Pixyz: a framework for developing complex deep generative models2019
- Author(s)
  Masahiro Suzuki
- Organizer
  Workshop on Deep Probabilistic Generative Models for Cognitive Architecture in Robotics (IROS2019)
- Invited / Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] Pixyz：複雑な深層生成モデル開発のためのフレームワーク2019
- Author(s)
  鈴木雅大, 金子貴輝, 谷口尚平, 松嶋達也, 松尾豊
- Organizer
  2019年度人工知能学会全国大会
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] UVTON: UV Mapping to Consider the 3D Structure of a Human in Image-Based Virtual Try-On Network2019
- Author(s)
  Shizuma Kubo, Yusuke Iwasawa, Masahiro Suzuki, Yutaka Matsuo
- Organizer
  Workshop on Computer Vision for Fashion, Art and Design, The IEEE International Conference on Computer Vision (ICCV 2019)
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] 身体の3次元構造を考慮したニューラル仮想試着2019
- Author(s)
  久保静真, 岩澤有祐, 鈴木雅大, 松尾豊
- Organizer
  2019年度人工知能学会全国大会
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] 深層生成モデルと世界モデル2019
- Author(s)
  鈴木雅大
- Organizer
  第4回統計・機械学習若手シンポジウム
- Data Source
  KAKENHI-PROJECT-19K21527
[Presentation] 半教師ありマルチモーダル深層生成モデルにおける共有表現の有効性と単一モダリティ入力への拡張2018
- Author(s)
  鈴木雅大，松尾豊
- Organizer
  2018年度人工知能学会全国大会
- Data Source
  KAKENHI-PROJECT-19K21527

1. 小寺聡 (80794776)

# of Collaborated Projects: 1 results

# of Collaborated Products: 1 results
2. 関倫久 (30528873)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results
3. 熊谷亘 (20747167)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results
4. 谷口忠大

# of Collaborated Projects: 0 results

# of Collaborated Products: 1 results

suzuki masahiro 鈴木 雅大

Suzuki Masahiro 鈴木 雅大

Research Projects

Research Products

Co-Researchers

タスク駆動による潜在空間選択機構を持つ世界モデルの構築Principal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

個別化医療の実現を目指したマルチモーダル汎用モデル開発

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

A Study on Deep Generative Models based on The Free Energy PrinciplePrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

A study on shared representation learning considering the uncertainty of each modalityPrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

[Journal Article] GIRL: Reward Function Learning Framework Independent of Text Generator Samples for Reinforcement Learning in Text Generation Tasks2024

Author(s)

Journal Title

DOI

ISSN

Year and Date

Language

Data Source

[Journal Article] Symbol Emergence and Representation Learning by Integrating GPLVM and Neural Network2024

Author(s)

Journal Title

DOI

ISSN

Language

Data Source

[Journal Article] Development and analysis of medical instruction-tuning for Japanese large language models2024

Author(s)

Journal Title

DOI

Data Source

[Journal Article] A Method for Detecting DeepFake Videos Using Angular Information of Faces2023

Author(s)

Journal Title

DOI

ISSN

Year and Date

Language

Data Source

[Journal Article] Learning global spatial information for multi-view object-centric models2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] World models and predictive coding for cognitive and developmental robotics: frontiers and challenges2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] 自由エネルギー原理と深層学習—世界モデルを軸として—2023

Author(s)

Journal Title

DOI

ISSN

Year and Date

Language

Data Source

[Journal Article] Scene Interpretation by Deep Generative Model Utilizing Information of Backgrounds2023

Author(s)

Journal Title

DOI

ISSN

suzuki masahiro 鈴木雅大

Suzuki Masahiro 鈴木雅大