Takashi Ishida

24 papers A* 5A 1Journal 17Unranked 1

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards. CoRR Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, Masashi Sugiyama
2025	J	`jnl`	How Can I Publish My LLM Benchmark Without Giving the True Answers Away? CoRR Takashi Ishida, Thanawat Lodkaew, Ikko Yamane
2025	J	`jnl`	Importance Weighting for Aligning Language Models under Deployment Distribution Shift. Trans. Mach. Learn. Res. Thanawat Lodkaew, Tongtong Fang, Takashi Ishida, Masashi Sugiyama
2025	J	`jnl`	LLM Routing with Dueling Feedback. CoRR Chao-Kai Chiang, Takashi Ishida, Masashi Sugiyama
2025	J	`jnl`	Off-Policy Corrected Reward Modeling for Reinforcement Learning from Human Feedback. CoRR Johannes Ackermann, Takashi Ishida, Masashi Sugiyama
2025	J	`jnl`	Practical estimation of the optimal classification error with soft labels and calibration. CoRR Ryota Ushio, Takashi Ishida, Masashi Sugiyama
2025	J	`jnl`	Scalable Oversight via Partitioned Human Supervision. CoRR Ren Yin, Takashi Ishida, Masashi Sugiyama
2024	A*	`conf`	Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical. ICML Wei Wang, Takashi Ishida, Yu-Jie Zhang, Gang Niu, Masashi Sugiyama
2023	J	`jnl`	Flooding Regularization for Stable Training of Generative Adversarial Networks. CoRR Iu Yahiro, Takashi Ishida, Naoto Yokoya
2023	A*	`conf`	Is the Performance of My Deep Network Too Good to Be True? A Direct Approach to Estimating the Bayes Error in Binary Classification. ICLR Takashi Ishida, Ikko Yamane, Nontawat Charoenphakdee, Gang Niu, Masashi Sugiyama
2023	J	`jnl`	Learning with Complementary Labels Revisited: A Consistent Approach via Negative-Unlabeled Learning. CoRR Wei Wang, Takashi Ishida, Yu-Jie Zhang, Gang Niu, Masashi Sugiyama
2023	A	`conf`	Mediated Uncoupled Learning and Validation with Bregman Divergences: Loss Family with Maximal Generality. AISTATS Ikko Yamane, Yann Chevaleyre, Takashi Ishida, Florian Yger
2022	J	`jnl`	Is the Performance of My Deep Network Too Good to Be True? A Direct Approach to Estimating the Bayes Error in Binary Classification. CoRR Takashi Ishida, Ikko Yamane, Nontawat Charoenphakdee, Gang Niu, Masashi Sugiyama
2022	J	`jnl`	Learning from Noisy Complementary Labels with Robust Loss Functions. IEICE Trans. Inf. Syst. Hiroki Ishiguro, Takashi Ishida, Masashi Sugiyama
2022	J	`jnl`	LocalDrop: A Hybrid Regularization for Deep Neural Networks. IEEE Trans. Pattern Anal. Mach. Intell. Ziqing Lu, Chang Xu, Bo Du, Takashi Ishida, Lefei Zhang, Masashi Sugiyama
2021	J	`jnl`	LocalDrop: A Hybrid Regularization for Deep Neural Networks. CoRR Ziqing Lu, Chang Xu, Bo Du, Takashi Ishida, Lefei Zhang, Masashi Sugiyama
2020	A*	`conf`	Do We Need Zero Training Loss After Achieving Zero Training Error? ICML Takashi Ishida, Ikko Yamane, Tomoya Sakai, Gang Niu, Masashi Sugiyama
2020	J	`jnl`	Do We Need Zero Training Loss After Achieving Zero Training Error? CoRR Takashi Ishida, Ikko Yamane, Tomoya Sakai, Gang Niu, Masashi Sugiyama
2019	A*	`conf`	Complementary-Label Learning for Arbitrary Losses and Models. ICML Takashi Ishida, Gang Niu, Aditya Krishna Menon, Masashi Sugiyama
2018	A*	`conf`	Binary Classification from Positive-Confidence Data. NeurIPS Takashi Ishida, Gang Niu, Masashi Sugiyama
2018	J	`jnl`	Complementary-Label Learning for Arbitrary Losses and Models. CoRR Takashi Ishida, Gang Niu, Aditya Krishna Menon, Masashi Sugiyama
2017	J	`jnl`	Binary Classification from Positive-Confidence Data. CoRR Takashi Ishida, Gang Niu, Masashi Sugiyama
2017	—	`conf`	Learning from Complementary Labels. NIPS Takashi Ishida, Gang Niu, Weihua Hu, Masashi Sugiyama
2017	J	`jnl`	Learning from Complementary Labels. CoRR Takashi Ishida, Gang Niu, Masashi Sugiyama