RIKEN Center for Advanced Intelligence Project AI Security and Privacy Team
Team Leader: Jun Sakuma (D.Eng.)
Research Summary
Along with the development of artificial intelligence (AI) technologies, decision making by machine learning is being widely implemented in society. In order for artificial intelligence to play a prominent role in decision making of humans, it is necessary to guarantee that private or confidential information is not leaked through the process of decision making by AI. Also, we need guarantee that that AI's behavior cannot be arbitrarily manipulated by a malicious attacker and AI cannot harm humans against the original AI designers’ intent. Our team aims to develop fundamental technologies of security and privacy that are necessary for proper use of AI in society.
Research Subjects:
- Artificial Intelligence Security
- Privacy Preservation in Artificial Intelligence
Main Research Fields
- Informatics
Related Research Fields
- Engineering
- Humanities
- Intelligent informatics
- Information security
Keywords
- machine learning
- security
- privacy
- fairness
Selected Publications
- 1.
Akifumi Wachi, Thien Q. Tran, Rei Sato, Takumi Tanabe, Youhei Akimoto:
"Stepwise Alignment for Constrained Language Model Policy Optimization."
NeurIPS 20246 - 2.
Yuwei Sun, Hideya Ochiai:
"Bidirectional Contrastive Split Learning for Visual Question Answering."
AAAI 2024: 21602-216096 - 3.
Jiayang Liu, Weiming Zhang, Kazuto Fukuchi, Youhei Akimoto, Jun Sakuma:
"Unauthorized AI cannot recognize me: Reversible adversarial example."
Pattern Recognit. 134: 109048 (2023)6 - 4.
Thien Q. Tran, Kazuto Fukuchi, Youhei Akimoto, Jun Sakuma:
"Statistically Significant Pattern Mining With Ordinal Utility."
IEEE Trans. Knowl. Data Eng. 35(9): 8770-8783 (2023)6 - 5.
Yoshimasa Akimoto, Kazuto Fukuchi, Youhei Akimoto, Jun Sakuma:
"Privformer: Privacy-preserving Transformer with MPC."
EuroS&P 2023: 392-4106 - 6.
Kaiwen Xu, Kazuto Fukuchi, Youhei Akimoto,Jun Sakuma:
"Statistically Significant Concept-based Explanation of Image Classifiers via Model Knockoffs."
IJCAI 2023: 519-5266 - 7.
Kazuto Fukuchi, Jun Sakuma:
"Demographic Parity Constrained Minimax Optimal Regression under Linear Model."
NeurIPS 20236 - 8.
Thien Q. Tran, Kazuto Fukuchi, Youhei Akimoto, Jun Sakuma:
"Unsupervised Causal Binary Concepts Discovery with VAE for Black-Box Model Explanation."
AAAI 2022: 9614-96226 - 9.
Takumi Tanabe, Rei Sato, Kazuto Fukuchi, Jun Sakuma, Youhei Akimoto:
"Max-Min Off-Policy Actor-Critic Method Focusing on Worst-Case Robustness to Model Misspecification."
NeurIPS 20226 - 10.
Rei Sato, Jun Sakuma, Youhei Akimoto:
"AdvantageNAS: Efficient Neural Architecture Search with Credit Assignment."
AAAI 2021: 9489-9496
Related Links
Lab Members
Principal investigator
- Jun Sakuma
- Team Leader
Core members
- Jun Seita
- Research Scientist
- Lu Sun
- Research Scientist
- Zhe Yu
- Postdoctoral Researcher
- Kazuto Fukuchi
- Visiting Scientist
- Hideitsu Hino
- Visiting Scientist
- Tatsuya Mori
- Visiting Scientist
- Yohei Akimoto
- Visiting Scientist
- Yuki Koike
- Visiting Scientist
- Takao Murakami
- Visiting Scientist
- Yuwei Sun
- Visiting Scientist
- Daiki Nishiyama
- Research Part-time Worker I
- Ziling He
- Research Part-time Worker I
- Junhao Wei
- Research Part-time Worker I
- Mikoto Kudou
- Research Part-time Worker I
- Takaaki Toda
- Research Part-time Worker II
- Shiwen An
- Research Part-time Worker II
- Md Ragib Amin Nihal
- Research Part-time Worker II
- Yu-Lin Tsai
- Student Trainee
- Takaki Kato
- Student Trainee
Contact Information
SB1006
1-1-1, Tennohdai, Tsukuba, Ibaraki
Email: jun.sakuma@riken.jp