Akira Horiguchi (he/him)

I am a Visiting Assistant Professor at the Department of Statistics at University of California, Davis.

• Previously I was a Postdoctoral Associate of the Department of Statistical Science at Duke University, working with Professors Li Ma and Cliburn Chan.
• I obtained my Ph.D. in Statistics from The Ohio State University, advised by Professors Matthew T. Pratola and Thomas J. Santner.
• I obtained my B.S. in Mathematics from the University of Maryland.

Here are my CV and other identifiers: Research Gate, orcid, Scholars@Duke, headshot, short bio.

My research interests are motivated by the modern statistical challenges in scientific applications such as climate science, computer experiments, and biomedical science, and involve nonparametric statistical approaches that quantify uncertainty in a principled way. Specific research areas include:

• Multi-level study designs: minimax rates, adaptive estimation, application to biomedical science
• Bayesian nonparametrics: stick-breaking models, model-based clustering, covariate-dependent mixture models, application to flow cytometry
• Computer experiments: sensitivity analysis, Pareto optimization, Bayesian regression trees, uncertainty quantification

Selected works:

Efficient Decision Trees for Tensor Regressions

H. Luo, A. Horiguchi, L. Ma.

Sampling depth trade-off in function estimation under a two-level design

A. Horiguchi, L. Ma, and B. T. Szabó.

Estimating Shapley Effects in Big-Data Emulation and Regression Settings using Bayesian Additive Regression Trees

A. Horiguchi and M. T. Pratola.

A tree perspective on stick-breaking models in covariate-dependent mixtures

A. Horiguchi, C. Chan, and L. Ma. Bayesian Analysis. To appear.

Using BART for Perform Pareto Optimization and Quantify its Uncertainties (2022) with arXiv and code

A. Horiguchi, T. J. Santner, Y. Sun, and M. T. Pratola. Technometrics, Special Issue on Industry 4.0.

Assessing variable activity for Bayesian regression trees (2021) with arXiv and code

A. Horiguchi, M. T. Pratola, and T. J. Santner. Reliability Engineering & Safety System, Special Issue on Sensitivity Analysis of Model Outputs.

Below are (free!) tools I've found useful.