Research topics (日/Eng)

私はこれまで以下のような研究を行ってきました。

• ベイズの理論と実践

• 地震学・測地学におけるデータ科学的手法開発

• 新・多変量データ解析

• 非標準データ解析 (多様体データ解析・ランキングデータ解析)

• 高次元データ解析

ベイズの理論と実践

[15] Akifumi Okuno and Keisuke Yano, A generalization gap estimation for overparameterized models  via Langevin functional variance,  Journal of Computational and Graphical  Statistics, vol. 0, 1061-8600, 2023 (https://doi.org/10.1080/10618600.2023.2197488; arXiv:2112.03660)

[14] Yukito Iba and Keisuke Yano, Posterior Covariance Information Criterion for Weighted Inference, accepted at Neural Computation, vol. 35, 1340–1361, 2023 (https://doi.org/10.1162/neco_a_01592,  arXiv:2106.13694)

[8] Edward George, Gourab Mukherjee, and Keisuke Yano, Optimal Shrinkage Estimation of Predictive Densities under  $\alpha$-divergences, Bayesian Analysis, vol.16, 1139-1155, 2021 (https://doi.org/10.1214/21-BA1264)

[7] Keisuke Yano, Ryoya Kaneko, and Fumiyasu Komaki, Minimax Predictive Density for Sparse Count Data,  Bernoulli, vol. 27, 1212-1238, 2021 (doi:10.3150/20-BEJ1271 ; python code)

[5] Keisuke Yano and Kengo Kato, On frequentist coverage errors of Bayesian credible sets in moderately high dimensions , Bernoulli, vol. 26, pp.616-641, 2020 (arXiv:1803.03450 ; doi: 10.3150/19-BEJ1142 )

[3] Keisuke Yano and Fumiyasu Komaki, Asymptotically minimax prediction in infinite sequence models, Electronic Journal of Statistics, vol. 11,pp. 3165-3195, 2017 (doi:10.1214/17-EJS1312).

[2] Keisuke Yano and Fumiyasu Komaki, Information criteria for prediction when the distributions of current and future observations differ, Statistica Sinica, vol. 27, pp. 1205-1223, 2017 (doi:10.5705/ss202015.0380).

[1] Keisuke Yano and Fumiyasu Komaki, Asymptotically constant-risk predictive densities when the distributions of data and target variables are different, Entropy, vol. 16, pp. 3026-3048, 2014 (doi:10.3390/e16063026).

地震学・測地学におけるデータ科学的手法開発

[12] Keisuke Yano and Masayuki Kano, l1 Trend Filtering-based Detection of Short-term Slow Slip Events: Application to a GNSS Array in Southwest Japan, Journal of Geophysical Research: Solid Earth, vol. 127, 5, e2021JB023258, 2022 (https://doi.org/10.1029/2021JB023258;python code)

[11] Yohta Yamanaka, Sumito Kurata, Keisuke Yano, Fumiyasu Komaki, Takahiro Shiina, Aitaro KatoStructured regularization based local earthquake tomography for the adaptation to velocity discontinuities, Earth, Planets and Space, vol. 74, 43, 2022 (https://doi.org/10.1186/s40623-022-01600-x)

[10] Hidenobu Takahashi, Kazuya Tateiwa, Keisuke Yano, Masayuki Kano, A convolutional neural network-based classification of local earthquakes and tectonic tremors in Sanriku-oki, Japan, using S-net data, Earth, Planets and Space, vol. 73, 186, 2021 (https://doi.org/10.1186/s40623-021-01524-y)

[9] Keisuke Yano, Takahiro Shiina, Sumito Kurata, Aitaro Kato, Fumiyasu Komaki, Shin'ichi Sakai,  and Naoshi Hirata, Graph-partitioning based convolutional network for earthquake detection using a seismic array, Journal of Geophysical Research: Solid Earth, vol. 126, e2020JB020269, 2021 (https://doi.org/10.1029/2020JB020269) Top downloaded article (2023/3/30) among work published in an issue between 1 Jan. 2021-31 Dec. 2021.

新・多変量データ解析

[16] Tomonari Sei and Keisuke Yano, Minimum information dependence modeling, accepted at Bernoulli (arXiv:2206.06792)

[13] Akifumi Okuno and Keisuke Yano, Dependence of variance on covariate design in nonparametric link regression, Statistics and Probability Letters, vol. 193,2023, 109716 (https://doi.org/10.1016/j.spl.2022.109716; arXiv:2012.13106)

非標準データ解析 (多様体データ解析・ランキングデータ解析)

[16] Tomonari Sei and Keisuke Yano, Minimum information dependence modeling, accepted at Bernoulli (arXiv:2206.06792)

[6] Kento Nakamura, Keisuke Yano, and Fumiyasu Komaki,  Adjacency-based regularization for partially ranked data with non-ignorable missing, Computational Statistics & Data Analysis, vol. 145, 106905, 2020 (doi:10.1016/j.csda.2019.106905)

[4] Yuya Takasu, Keisuke Yano, and Fumiyasu Komaki, Scoring Rules for Statistical Models on Spheres, Statistics and Probability Letters, vol. 138, pp. 111-115, 2018 (doi:10.1016/j.spl.2018.02.054).

高次元データ解析

[15] Akifumi Okuno and Keisuke Yano, A generalization gap estimation for overparameterized models  via Langevin functional variance,  Journal of Computational and Graphical  Statistics, vol. 0, 1061-8600, 2023 (https://doi.org/10.1080/10618600.2023.2197488; arXiv:2112.03660)

[12] Keisuke Yano and Masayuki Kano, l1 Trend Filtering-based Detection of Short-term Slow Slip Events: Application to a GNSS Array in Southwest Japan, Journal of Geophysical Research: Solid Earth, vol. 127, 5, e2021JB023258, 2022 (https://doi.org/10.1029/2021JB023258;python code)

[11] Yohta Yamanaka, Sumito Kurata, Keisuke Yano, Fumiyasu Komaki, Takahiro Shiina, Aitaro KatoStructured regularization based local earthquake tomography for the adaptation to velocity discontinuities, Earth, Planets and Space, vol. 74, 43, 2022 (https://doi.org/10.1186/s40623-022-01600-x)

[7] Keisuke Yano, Ryoya Kaneko, and Fumiyasu Komaki, Minimax Predictive Density for Sparse Count Data,  Bernoulli, vol. 27, 1212-1238, 2021 (doi:10.3150/20-BEJ1271 ; python code)

[5] Keisuke Yano and Kengo Kato, On frequentist coverage errors of Bayesian credible sets in moderately high dimensions , Bernoulli, vol. 26, pp.616-641, 2020 (arXiv:1803.03450 ; doi: 10.3150/19-BEJ1142 )

[3] Keisuke Yano and Fumiyasu Komaki, Asymptotically minimax prediction in infinite sequence models, Electronic Journal of Statistics, vol. 11,pp. 3165-3195, 2017 (doi:10.1214/17-EJS1312).