Matt Holland :: Info

I am studying and working as a JSPS Research Fellow (DC2) at the Mathematical Informatics Lab at NAIST in Ikoma, Japan.


Work  |  Teaching  |  Contact  |  Misc.


Research focus

In a broad sense, my interests concentrate around the theoretical foundations of machine learning, and the ways in which this kind of very precise knowledge can be leveraged to design reliable statistical estimation procedures. That is, "algorithms that work for reasons we understand."


Selected publications and related materials

Efficient learning with robust gradient descent.
Preprint.
Matthew J. Holland and Kazushi Ikeda.
[arXiv]

Robust regression using biased objectives.
Matthew J. Holland and Kazushi Ikeda.
European Conference on Machine Learning & Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2017.
Skopje, Macedonia
[pdf, code]

Minimum proper loss estimators for parametric models.
Matthew J. Holland and Kazushi Ikeda.
IEEE Transactions on Signal Processing, 64(3):704-713, 2016.
[pdf, data]

Stabilization of learning algorithms and optimal suppression of noise.
Matthew J. Holland and Kazushi Ikeda.
30th Annual Conference of the Japan Society of Artificial Intelligence, 2016.
Kitakyushu, Japan

Learning volatile targets via strategic truncation.
Matthew J. Holland and Kazushi Ikeda.
18th Information-Based Induction Sciences Workshop (IBIS), 2015.
Tsukuba, Japan

Location robust estimation of predictive Weibull parameters in short-term wind speed forecasting.
Matthew J. Holland and Kazushi Ikeda.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015.
Brisbane, Australia

Density estimation using proper loss functions.
Matthew J. Holland and Kazushi Ikeda.
IEICE Signal Processing Workshop, 2015.
Ishigakijima, Japan

A probabilistic forecasting approach to wind turbine control.
Matthew J. Holland and Kazushi Ikeda.
SICE Journal of Control, Measurement, and System Integration, 8(1):61-66, 2015.
[pdf]

Very short term predictive wind turbine control.
Matthew J. Holland and Kazushi Ikeda.
SICE System and Information Division Annual Event (SSI), 2014.
Okayama, Japan

Forecasting in wind energy applications with site-adaptive Weibull estimation.
Matthew J. Holland and Kazushi Ikeda.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2014.
Florence, Italy

Short-term probabilistic wind forecasting and data-driven decision making in electricity markets.
Matthew J. Holland and Kazushi Ikeda.
IEICE 28th Signal Processing Symposium, 2013.
Shimonoseki, Japan

A probabilistic forecasting approach to wind turbine control.
Matthew J. Holland and Kazushi Ikeda.
SICE System and Information Division Annual Event (SSI), 2013.
Otsu, Japan


Teaching

My enthusiasm for education, on both the giving and receiving ends, is very strong. I have found that the satisfaction gained after completing a strenuous research project can easily be matched by a short seminar including some dedicated Q&A time with students.

  • Introduction to Programming (KWU, 2016-2017, 2017-2018)

  • Linear Algebra (NAIST, 2015) [ref]

  • Bioinformatics seminar (NMU, 2015)


Contact

I can be reached at the following address:


Miscellany

In the past, I studied microeconomic theory and more practitioner-oriented finance. Having lived in Kyoto, Osaka and Nara for some time, I have also put a fair amount of effort into trying to achieve fluency in written and spoken Japanese. Now, besides research-related work, I spend most of my free time cycling, reading, or eating/drinking in Osaka.