ホーランド マシュー

Making Robust Generalizers Less Rigid with Loss Concentration 

Proceedings of ICONIP 2025, Communications in Computer and Information Science 2754 391-405 2026年1月  査読有り

Learning as choosing a loss distribution 

IEEE VIS 2025, Workshop on Visualization for AI Explainability (VISxAI) 2025年11月  査読有り

Soft ascent-descent as a stable and flexible alternative to flooding 

Advances in Neural Information Processing Systems 37 (NeurIPS 2024) 37955-37987 2024年12月  査読有り

Criterion Collapse and Loss Distribution Control 

Proceedings of Machine Learning Research (ICML 2024) 235 18547-18567 2024年7月  査読有り

Robust variance-regularized risk minimization with concomitant scaling 

Proceedings of Machine Learning Research (AISTATS 2024) 238 1144-1152 2024年5月  査読有り

A Survey of Learning Criteria Going Beyond the Usual Risk 

Journal of Artificial Intelligence Research 78 781-821 2023年11月  査読有り

Flexible risk design using bi-directional dispersion 

Proceedings of Machine Learning Research (AISTATS 2023) 206 1586-1623 2023年4月  査読有り

Learning with risks based on M-location 

Machine Learning 111 4679-4718 2022年8月  査読有り

Spectral risk-based learning using unbounded losses 

Proceedings of Machine Learning Research (AISTATS 2022) 151(1) 1871-1886 2022年3月  査読有り

Anytime Guarantees under Heavy-Tailed Data 

Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2022) 36(6) 6918-6925 2022年2月  査読有り

Making learning more transparent using conformalized performance prediction 

ICML 2021, Workshop on Distribution-Free Uncertainty Quantification 2021年7月  査読有り

Robustness and scalability under heavy tails, without strong convexity 

Proceedings of Machine Learning Research (AISTATS 2021) 130(1) 2021年4月  査読有り

Learning with risk-averse feedback under potentially heavy tails 

Proceedings of Machine Learning Research (AISTATS 2021) 130(1) 2021年4月  査読有り

Scaling-Up Robust Gradient Descent Techniques 

Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2021) 35(9) 7694-7701 2021年2月  査読有り

Better scalability under potentially heavy-tailed feedback 

Archival version 2020年12月  

PAC-Bayes under potentially heavy tails 

Advances in Neural Information Processing Systems 32 (NeurIPS 2019) 2019年12月  査読有り

Efficient learning with robust gradient descent 

Machine Learning 108(8) 1523-1560 2019年9月  査読有り

Distribution-robust mean estimation via smoothed random perturbations 

arXiv preprint 2019年6月  

Better generalization with less data using robust gradient descent 

Proceedings of Machine Learning Research (ICML 2019) 97 2761-2770 2019年6月  査読有り

Robust descent using smoothed multiplicative noise 

Proceedings of Machine Learning Research (AISTATS 2019) 89(1) 703-711 2019年4月  査読有り