革新知能統合研究センター

• Attacks Which Do Not Kill Training Make Adversarial Learning Stronger
  Jingfeng Zhang (National University of Singapore)*
  Xu Xilie (Shandong University)
  Bo Han (HKBU / RIKEN AIP)
  Gang Niu (RIKEN AIP)
  Lizhen Cui (ShanDong University)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
  Mohan Kankanhalli (National University of Singapore)
• Variational Imitation Learning with Diverse-quality Demonstrations
  Voot Tangkaratt (RIKEN AIP)
  Bo Han (HKBU / RIKEN AIP)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• SIGUA: Forgetting May Make Learning with Noisy Labels More Robust
  Bo Han (HKBU / RIKEN AIP)
  Gang Niu (RIKEN AIP)
  Xingrui Yu (University of Technology Sydney)
  Quanming Yao (4Paradigm)
  Miao Xu (University of Queensland/ RIKEN AIP)
  Ivor Tsang (University of Technology Sydney)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Online Dense Subgraph Discovery via Blurred-Graph Feedback
  Yuko Kuroki (The University of Tokyo / RIKEN AIP)
  Atsushi Miyauchi (University of Tokyo / RIKEN AIP)
  Junya Honda (University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Few-shot Domain Adaptation by Causal Mechanism Transfer
  Takeshi Teshima (The University of Tokyo / RIKEN AIP)
  Issei Sato (University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Fast Deterministic CUR Matrix Decomposition with Accuracy Assurance
  Yasutoshi Ida (NTT)
  Sekitoshi Kanai (NTT Software Innovation Center)
  Yasuhiro Fujiwara (NTT Communication Science Laboratories)
  Tomoharu Iwata (NTT)
  Koh Takeuchi (NTT)
  Hisashi Kashima (Kyoto University / RIKEN AIP)
• Learning with Multiple Complementary Labels
  Lei Feng (Nanyang Technological University)*
  Takuo Kaneko (The University of Tokyo / RIKEN AIP)
  Bo Han (HKBU / RIKEN AIP)
  Gang Niu (RIKEN AIP)
  Bo An (Nanyang Technological University)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Evolutionary Topology Search for Tensor Network Decomposition
  Chao Li (RIKEN AIP)
  Zhun Sun (RIKEN AIP)
• Accelerating the diffusion-based ensemble sampling by non-reversible dynamics
  Futoshi Futami (The University of Tokyo/ RIKEN AIP)
  Issei Sato (University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Unbiased Risk Estimators Can Mislead: A Case Study of Learning with Complementary Labels
  Yu-Ting Chou (National Taiwan University)*
  Gang Niu (RIKEN AIP)
  Hsuan-Tien Lin (National Taiwan University)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Searching to Exploit Memorization Effect in Learning with Noisy Labels
  Quanming Yao (4Paradigm)
  Hansi Yang (Tsinghua)
  Bo Han (HKBU / RIKEN AIP)
  Gang Niu (RIKEN AIP)
  James Kwok (Hong Kong University of Science and Technology)
• Normalized Flat Minima: Exploring Scale Invariant Definition of Flat Minima for Neural Networks Using PAC-Bayesian Analysis
  Yusuke Tsuzuku (The University of Tokyo / RIKEN AIP)
  Issei Sato (University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Progressive Identification of True Labels for Partial-Label Learning
  Jiaqi Lv (Southeast University)*
  Miao Xu (University of Queensland/ RIKEN AIP)
  Lei Feng (Nanyang Technological University)*
  Gang Niu (RIKEN AIP)
  Xin Geng (Southeast University)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Graph Homomorphism Convolution
  Hoang Nguyen (RIKEN AIP)
  Takanori Maehara (RIKEN AIP)
• Multi-fidelity Bayesian Optimization with Max-value Entropy Search and its Parallelization
  Shion Takeno (Nagoya Institute of Technology)
  Hitoshi Fukuoka (Nagoya University)
  Yuhki Tsukada (Nagoya University)
  Toshiyuki Koyama (Nagoya University)
  Motoki Shiga (Gifu University)
  Ichiro Takeuchi (Nagoya Institute of Technology / RIKEN AIP)
  Masayuki Karasuyama (Nagoya Institute of Technology)
• Do We Need Zero Training Loss After Achieving Zero Training Error?
  Takashi Ishida (The University of Tokyo / RIKEN AIP)
  Ikko Yamane (The University of Tokyo)
  Tomoya Sakai (NEC)
  Gang Niu (RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Handling the Positive-Definite Constraint in the Bayesian Learning Rule
  Wu Lin (UBC)+
  Mark Schmidt (University of British Columbia)
  Mohammad Emtiyaz Khan (RIKEN AIP)
• Training Binary Neural Networks using the Bayesian Learning Rule
  Xiangming Meng (RIKEN AIP)
  Roman Bachmann (EPFL)*
  Mohammad Emtiyaz Khan (RIKEN AIP)

関連研究室