Hiroshi Nakagawa

J-GLOBAL         Last updated: Jul 5, 2016 at 14:22
 
Avatar
Name
Hiroshi Nakagawa
Affiliation
The University of Tokyo
Section
Information Technology Center
Twitter ID
@hiroshnakagawa3

Research Areas

 
 

Published Papers

 
Regret Lower Bound and Optimal Algorithm in Dueling Bandit Problem
Junpei Komiyama, Junya Honda, Hisashi Kashima and Hiroshi Nakagawa
COLT2015: Conference On Learning Theory   1141-1154   Jul 2015   [Refereed]
Optimal Regret Analysis of Thompson Sampling in Stochastic Multi-armed Bandit Problem with Multiple Plays
Junpei Komiyama, Junya Honda, and Hiroshi Nakagawa
ICML2015: International Conference on Machine Learning   1152-1161   Jul 2015   [Refereed]
Bayesian Differential Privacy on Correlated Data
Bing Yang, Issei Sato, Hiroshi Nakagawa
ACM SIGMOD2015   747-762   Jun 2015   [Refereed]
Robust Distributed Training of Linear Classifiers Based on Divergence Minimization Principle
Junpei Komiyama, Hidekazu Ooiwa, Hiroshi Nakagawa
ECML PKDD 2014:Springer Lecture Notes Artificial Intelligence   LNAI(8725) 1-17   Sep 2014   [Refereed]
The Hybrid Nested Hierarchical Dirichlet Process and Its Application to Topic Modeling with Word Differentiation
Tengfei Ma, Issei Sato and Hiroshi Nakagawa
AAAI2015      Jan 2015   [Refereed]
Special Features:Towrad Fair Use of the Personal Data: Foreword
Hiroshi Nakagawa
IPSJ Magazine   55(12) 1332-1336   Dec 2014
Online and Stochastic Learning with a Human Cognitive Bias
Hidekazu Oiwa, Issei Sato,Hiroshi Nakagawa
AAAI2014   2020-2026   Jun 2014   [Refereed]
Automatically Determining a Proper Length for Multi-document Summarization: A Bayesian Nonparametric Approach
Tengfei Ma, Hiroshi Nakagawa
ACL SIGDAT Conference of Emprical Methods on Natural Language Processing   749-736   Dec 2013   [Refereed]
Probablistic Bandit Constrained by Lock-up Period
Junpei Komiyama, Hiroshi Nakagawa
IPSJ TOM   6(3) 11-22   Dec 2013   [Refereed]
Healing Truncation Bias : Self-weighted Truncation framework for Dual Averaging
Hidekazu Ooiwa, Shin Matsushima, Hiroshi Nakagawa
IEEE International Conference on Data Mining   2012    Dec 2012   [Refereed]
Practical Collapsed Variational Bayes Inference for Hierarchical Dirichlet Process
Issei Sato, Ken-ich Kurihara,Hiroshi Nakagawa
International Conference on Knowledge Discovery and Data Mining   KDD(2012) 105-113   Aug 2012   [Refereed]
Rethinking Collapsed Variational Bayes Inference for LDA
Issei Sato, Hiroshi Nakagawa
International Conference on Machine Learning (ICML 2012)   999-1006   Jul 2012   [Refereed]
Reducing Wrong Labels in Distant Supervision for Relation Extraction
Shingo Takamatsu,Hiroshi Nakagawa
Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics   ACL2012(1) 721-729   Jul 2012   [Refereed]
Yang Bin, Hiroshi Nakagawa
Springer Lecture Notes   7301 542-553   Jun 2012   [Refereed]
Yang Bin, Hiroshi Nakagawa
Springer Lecture Note   7301 542-553   May 2012   [Refereed]
Secure Clustering in Private Networks
Bin Yang, Hiroshi Nakagawa
11th IEEE International Conference on Data Mining(ICDM)   11 894-903   Dec 2011   [Refereed]
Probabilistic Frequency-aware Truncated methods for Sparse Online Learning
Hidekazu Ooiwa, Shin Matsushima, Hiroshi Nakagawa
Springer Lecture Notes Artificial Intelligence (LNAI) ECML PKDD ,   6911(Part II) 533-548   Sep 2011   [Refereed]
Probabilistic Matrix Factorization Leveraging Contexts for Unsupervised Relation Extraction
Shingo Takamatsu, Issei Sato, Hiroshi Nakagawa
Springer Lecture Notes Artificial Intelligence (LNAI)   6634(Part I) 87-99   May 2011   [Refereed]
Shibayama Naoki, Nakagawa Hiroshi
Transactions of the Japanese Society for Artificial Intelligence   26(1) 262-272   2011
probabilistic Latent Semantic Indexing (pLSI) is a fundamental method for the analysis of text and related resources which is based on a simple statistical model. This method has high extendibility and scalability due to its simplicity. pLSI is al...
Collusion-Resistant Privacy-Preserving Data Mining
Bin Yang, Hiroshi Nakagawa, Issei Sato, Jun Sakuma
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining   16th 483-492   Jul 2010   [Refereed]

Misc

 
SATO Issei, KURIHARA Kenichi, TANAKA Shu, MIYASITA Seiji, NAKAGAWA Hiroshi
IEICE technical report   111(87) 111-117   Jun 2011
We develope quantum annealing (QA) for the infinite mixture models. The QA is regarded as a parallelized ex-tension of simulated annealing (SA). Experiments show that QA finds a better seating arrangement than SA in the Chinese restaurant process....
Hidekazu Oiwa, Shin Matsushima, Hiroshi Nakagawa
IPSJ SIG Notes   2011(18) 1-6   Feb 2011
Online learning is a method that updates parameters whenever it receives a single data. Online learning can learn efficiently from large data set. L1 regularization is used for inducing sparsity into parameters and exclude unnecessary parameters. ...
SATO Issei, NAKAGAWA Hiroshi
IEICE technical report   110(76) 33-39   Jun 2010
Topic model is a probabilistic generative model that models latent semantics of words as "topic". In this paper, we propose a novel topic model that captures the power-law phenomenon of a word distribution, which is known as Zipf's law in linguist...
Yo Ehara, Takashi Ninomiya, Nobuyuki Shimizu, Hiroshi Nakagawa
IPSJ SIG Notes   2010(18) 1-7   May 2010
We have been proposing a system to support reading English Web pages for ESL (English as a second language) users by annotate the words they may not know by personalized prediction of those words using the word click logs. In this paper, we propos...
EHARA YO, NINOMIYA TAKASHI, SHIMIZU NOBUYUKI, NAKAGAWA HIROSHI
情報処理学会研究報告. HCI, ヒューマンコンピュータインタラクション研究会報告 = IPSJ SIG technical reports   138 1-7   May 2010
We have been proposing a system to support reading English Web pages for ESL (English as a second language) users by annotate the words they may not know by personalized prediction of those words using the word click logs. In this paper, we propos...
IKEDA Masaki, SATO Issei, YOSHIDA Minoru, Nakagawa Hiroshi
全国大会講演論文集   72(5) "5-109"-"5-110"   Mar 2010
Mori Tatsuya, Masuda Hidetaka, Nakagawa Hiroshi, Kiyota Yoji
全国大会講演論文集   72(5) "5-181"-"5-182"   Mar 2010
MAEDA Akira, NAKAGAWA Hiroshi
全国大会講演論文集 / 情報処理学会 [編]   72(1) 1-921   Mar 2010
Matsushima Shin, Shimizu Nobuyuki, Yoshida Kazuhiro, Ninomiya Takashi, Nakagawa Hiroshi
全国大会講演論文集   72(2) "2-467"-"2-468"   Mar 2010
多くの機械学習ではデータ全体に対して1つの最適な関数を求めるため,大量のデータに対して学習を行う際に計算コストの面から困難が生じる場合があるが,オンライン学習法は,データを1つずつ受け取りながら逐次的に識別関数を学習していく手法であり,低い計算コストで学習可能である.Crammerらによって提案されたPassive-Aggressiveアルゴリズム(PAアルゴリズム)は代表的なオンライン学習アルゴリズムであるが,多クラス識別問題においては,厳密なPAアルゴリズムの枠組みから外れた近似解法が...

Books etc

 
Machine Learning
Hiroshi Nakagawa
Nov 2015   ISBN:978-4-621-08991-0
Information Law
Hiroshi Nakagawa (Part:Joint Work)
Sep 2013   
長尾 真, 松本 裕治, 橋田 浩一, ジョン ベイトマン, 中川 裕志 (Part:Joint Work)
岩波書店   Nov 2004   ISBN:400006908X
共立出版   Mar 2003   ISBN:4320120639
Recent Advances in Computational Terminology
Hiroshi Nakagawa (Part:Joint Work, Experimental evaluation of ranking and selection methods in term extraction)
John Benjamins   Oct 2001   

Works

 
Statistical Machine Learning
Hiroshi Nakagawa   The Others   Apr 2005 - Today
Hiroshi Nakagawa   The Others   Jan 1990 - Today
Language Model
Hiroshi Nakagawa   The Others   Apr 1986 - Mar 1996
Logic Programming
Hiroshi Nakagawa   The Others   Apr 1983 - Mar 1993
Image Processing
Hiroshi Nakagawa   The Others   Apr 1975 - Mar 1982

Research Grants & Projects

 
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(A))
Project Year: 2009 - 2012    Investigator(s): Hiroshi NAKAGAWA
We developed a clustering system which makes clusters of web pages in response to a person name query in 2009 as planned, and evaluate it experimentally. In 2010, our contribution is a new non-negative probabilistic matrix decomposition algorithm ...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(B))
Project Year: 2008 - 2010    Investigator(s): Takehito UTSURO
In this project, given a collection of multilingual documents available on the Web, we cross-lingually and cross-culturally compare less well known facts and opinions that are observed in the collected documents. We conceptually index the space of...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(A))
Project Year: 2005 - 2007    Investigator(s): Hiroshi NAKAGAWA
We aimed at a system that extracts texts or part of texts including knowledge which various users are interested in from huge amount of Web pages in this research. We developed the following systems for this purpose.(1) A system which extracts ter...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(C))
Project Year: 2000 - 2001    Investigator(s): Hiroshi NAKAGAWA
We mainly grappled with automatic term extraction methods which extracts domain specific terms from domain corpora that were distributed by NTCIR1 TMREC task group. Among various works in automatic term extraction, the majority of them are concern...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(C))
Project Year: 1999 - 2000    Investigator(s): Tatsunori MORI
The purpose of this research is establishment of the method for "content"-based information retrieval. In our research, the "content" is regarded as the combination of the following items : a) Logical structure of document annotated by tags, b) Te...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(B))
Project Year: 1997 - 1998    Investigator(s): Hiroshi NAKAGAWA
A technical manual which is easy for a reader to find the part that the reader wants to know is desperately needed in this electronics age. This is not accomplished solely by the mere electronic manual. We still need many technical advancements fo...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(国際学術研究)
Project Year: 1994 - 1995    Investigator(s): Takao GUNJI
1994In August, the head investigator and some of the investigators visited Stanford University for about a week and had a series of discussions with the U.S.investigators. The Japanese side presented their previous studies on Japanese based on con...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(国際学術研究)
Project Year: 1994 - 1995    Investigator(s): Hiroshi NAKAGAWA
Through two years of this joint research project, we worked out the following results.1. Reciprocal : We did comparative study of English "each other" and Japanese "otagai" and "-aw" and end up with the conclusion that their interpretations depend...
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(一般研究(C))
Project Year: 1986 - 1987    Investigator(s): Hiroshi NAKAGAWA
The purpose of this research is to design a concepts learning system using nonmonotonic logic. We use TMS's depebdency directed backtracking and McCarthy's circumscription especially. We describe the out-line of out system in the following.1. Iden...