Rchr
J-GLOBAL ID:201701008049737949
Update date: Aug. 08, 2024
Holland Matthew J.
ホーランド マシュー | Holland Matthew J.
Affiliation and department:
Job title:
Assistant Professor
Research keywords (2):
Statistical learning theory
, Machine learning
Research theme for competitive and other funds (9):
- 2022 - 2026 Novel learning algorithms through off-sample generalization metric design
- 2021 - 2025 Machine learning with guarantees under diverse risk measures
- 2022 - 2022 Designing new learning algorithms via novel feedback generation mechanisms
- 2020 - 2022 Learning with guarantees under more diverse notions of risk
- 2019 - 2022 Robust and efficient learning algorithms through control of margin distributions
- 2019 - 2020 Stabilization and performance guarantees for machine learning methods via margin distribution control
- 2018 - 2019 Robust and efficient learning algorithms through control of margin distributions
- 2017 - 2019 Safe AI is efficient AI: improved generalization via robust learning algorithms
- 2016 - 2017 Analysis of automated loss-selecting algorithms and their extension to time-series data
Show all
Papers (24):
-
Matthew J. Holland, Toma Hamada. Making Robust Generalizers Less Rigid with Soft Ascent-Descent. arXiv preprint. 2024
-
Matthew J. Holland. Criterion Collapse and Loss Distribution Control. Proceedings of Machine Learning Research (ICML 2024). 2024. 235. 18547-18567
-
Matthew J. Holland. Robust variance-regularized risk minimization with concomitant scaling. Proceedings of Machine Learning Research (AISTATS 2024). 2024. 238. 1144-1152
-
Matthew J. Holland, Kazuki Tanabe. A Survey of Learning Criteria Going Beyond the Usual Risk. Journal of Artificial Intelligence Research. 2023. 78. 781-821
-
Matthew J. Holland, Kosuke Nakatani. Implicit regularization via soft ascent-descent. arXiv preprint. 2023
more...
MISC (2):
-
Matthew J. Holland. 5分で分かる!? 有名論文ナナメ読み:Duchi, J. et al. : Adaptive Subgradient Methods for Online Learning and Stochastic Optimization. IPSJ Magazine. 2019. 60. 8. 780-781
-
Matthew J. Holland. Robust gradient descent via back-propagation: A Chainer-based tutorial. 2019
Works (14):
-
addro: learning with concentrated OCE
Matthew J. Holland 2024 -
-
collapse: a study of criterion collapse in machine learning
Matthew J. Holland 2024 -
-
bdd-flood: links between bi-directional dispersion and flooding
Matthew J. Holland 2023 -
-
bdd-mv: mean-variance minimization using bi-directional dispersion functions
Matthew J. Holland 2023 -
-
offgen: a visual "explainer" for off-sample generalization metrics
Matthew J. Holland 2022 -
more...
Professional career (1):
- Ph.D (Nara Institute of Science and Technology)
Awards (8):
- 2024/08 - Transactions on Machine Learning Research (TMLR) Expert Reviewer Recognition (2024)
- 2023/07 - Transactions on Machine Learning Research (TMLR) Expert Reviewer Recognition (2023)
- 2022/07 - Japanese Neural Network Society Distinguished Presentation Award (NEURO2022)
- 2021/09 - Japanese Neural Network Society Young Researcher Award (JNNS2021)
- 2019/11 - IEICE, Information-Based Induction Sciences and Machine Learning (IBISML) Technical Committee Award finalist (IBIS 2019)
- 2018/03 - Foundation for NAIST NAIST Top Student Award (PhD Program)
- 2017/02 - IEEE Kansai Section Student Paper Award
- 2015/03 - Foundation for NAIST NAIST Top Student Award (Master's Program)
Show all
Return to Previous Page