Hiroshi NAGAOKA (長岡浩司)

0151700_名誉教授・その他関係者, 0150000_全学支援組織・その他, The University of Electro-Communications

ジャーナル論文 - rm_published_papers: International Conference Proceedings 査読済み

Information-geometrical characterization of statistical models which are statistically equivalent to probability simplexes

by Hiroshi NAGAOKA

公開済 06/2017

Proc. of 2017 IEEE International Symposium on Information Theory (ISIT), 1346 - 1350

ジャーナル論文 - rm_published_papers: International Conference Proceedings 査読済み

A Two-Step Lniversal VV Code L sing an Asymptotically Sufficient Statistic Extracted from a VF Code

by Hiroshi NAGAOKA

公開済 2016

PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016), Proc. of ISITA 2016, 6 - 10

ジャーナル論文 - rm_published_papers: International Conference Proceedings 査読済み

A characterization of statistical manifolds on which the relative entropy is a Bregman divergence

by Hiroshi NAGAOKA

公開済 2016

2016 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, 1656 - 1659

ジャーナル論文 - rm_published_papers: Scientific Journal 査読済み

A Fundamental Inequality for Lower-bounding the Error Probability for Classical and Classical-Quantum Multiple Access Channels and Its Applications,

by Hiroshi NAGAOKA

公開済 12/2015

IEICE Trans. Fundamentals, E98-A, 12, 2376 - 2383

ジャーナル論文 - rm_published_papers: International Conference Proceedings 査読済み

Lower Bounds on the Error Probability in Classical and Quantum State Discrimination

by Hiroshi NAGAOKA

公開済 10/2012

Proc. of ISITA2012, 299 - 302

ジャーナル論文 - rm_published_papers: Scientific Journal 査読済み

量子i.i.d.状態の仮説検定に関する数値的手法とその誤差分析

by Hiroshi NAGAOKA

公開済 10/2011

情報処理学会論文誌コンピューティングシステム (ACS), 4, 4, 214 - 227

本論文ではテンソル積の既約分解を用いた 2 × 2 サイズの密度行列の量子 i.i.d. 状態に対する仮説検定問題の数値的手法を扱う．特に，誤り確率の計算に現れる数値誤差について分析を行った．その結果，既約成分ごとの寄与の相対関係が重要であることを見い出した．また，多倍長桁数と固有値分解の収束判定に用いる ε を変えたときに，数値誤差がどのように現れるかを考察した．In this paper, we deal with a numerical method for hypothesis testing of quantum i.i.d. states for 2 × 2 density matrices based on the irreducible decomposition of tensor products. Especially, we analyze numerical errors appearing in computation of the error probability. As a result, we found out that the mutual relation among the contributions of irreducible elements is important. In addition, we studied how numerical error appears for different settings of multiple-precision and the convergence criterion ε of eigenvalue routine.

ジャーナル論文 - rm_published_papers: International Conference Proceedings 査読済み

A Numerical Study of Hypothesis Testing for Quantum I.I.D. States

by Hiroshi NAGAOKA

公開済 08/2010

Proc., 213 - 214