Masashi Sugiyama

45

Papers

36

Total Citations

Papers (45)

Towards Effective Evaluations and Comparisons for LLM Unlearning Methods

Direct Distillation between Different Domains

Robust Multi-View Learning via Representation Fusion of Sample-Level Attention and Alignment of Simulated Perturbation

Towards Out-of-Modal Generalization without Instance-level Modal Correspondence

The adaptive complexity of parallelized log-concave sampling

Dual-Decoupling Learning and Metric-Adaptive Thresholding for Semi-Supervised Multi-Label Learning

Action-Agnostic Point-Level Supervision for Temporal Action Detection

Thompson Sampling for Real-Valued Combinatorial Pure Exploration of Multi-Armed Bandit

The Choice of Noninformative Priors for Thompson Sampling in Multiparameter Bandit Models

Locally Estimated Global Perturbations are Better than Local Perturbations for Federated Sharpness-aware Minimization

A General Framework for Learning from Weak Supervision

Efficient Non-stationary Online Learning by Wavelets with Applications to Online Distribution Shift Adaptation

Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical

Balancing Similarity and Complementarity for Federated Learning

Counterfactual Reasoning for Multi-Label Image Classification via Patching-Based Training

Generating Chain-of-Thoughts with a Pairwise-Comparison Approach to Searching for the Most Promising Intermediate Thought

On Focal Loss for Class-Posterior Probability Estimation: A Theoretical Perspective

Instance-Dependent Label-Noise Learning With Manifold-Regularized Transition Matrix Estimation

Distribution Shift Matters for Knowledge Distillation with Webly Collected Images

Multi-Label Knowledge Distillation

Coupling-based Invertible Neural Networks Are Universal Diffeomorphism Approximators

NeurIPS 2020arXiv

Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learning

NeurIPS 2020arXiv

Part-dependent Label Noise: Towards Instance-dependent Label Noise

NeurIPS 2020arXiv

Learning from Aggregate Observations

NeurIPS 2020arXiv

Analysis and Design of Thompson Sampling for Stochastic Partial Monitoring

NeurIPS 2020arXiv

Provably Consistent Partial-Label Learning

NeurIPS 2020arXiv

Rethinking Importance Weighting for Deep Learning under Distribution Shift

NeurIPS 2020arXiv

Loss function based second-order Jensen inequality and its application to particle variational inference

NeurIPS 2021arXiv

Probabilistic Margins for Instance Reweighting in Adversarial Training

NeurIPS 2021arXiv

Generalizing Consistent Multi-Class Classification with Rejection to be Compatible with Arbitrary Losses

Learning Contrastive Embedding in Low-Dimensional Space

Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks

NeurIPS 2022arXiv

Adapting to Online Label Shift with Provable Guarantees

NeurIPS 2022arXiv

Synergy-of-Experts: Collaborate to Improve Adversarial Robustness

On the Overlooked Pitfalls of Weight Decay and How to Mitigate Them: A Gradient-Norm Perspective

NeurIPS 2023arXiv

Binary Classification with Confidence Difference

NeurIPS 2023arXiv

Distributional Pareto-Optimal Multi-Objective Reinforcement Learning

Enhancing Adversarial Contrastive Learning via Adversarial Invariant Regularization

NeurIPS 2023arXiv

Diversified Outlier Exposure for Out-of-Distribution Detection via Informative Extrapolation

NeurIPS 2023arXiv

Generalizing Importance Weighting to A Universal Solver for Distribution Shift Problems

NeurIPS 2023arXiv

Class-Distribution-Aware Pseudo-Labeling for Semi-Supervised Multi-Label Learning

Adapting to Continuous Covariate Shift via Online Density Ratio Estimation

NeurIPS 2023arXiv

Online (Multinomial) Logistic Bandit: Improved Regret and Constant Computation Cost

Imitation Learning from Vague Feedback

Efficient Adversarial Contrastive Learning via Robustness-Aware Coreset Selection

NeurIPS 2023arXiv