PROGRAM

Aug 3(Thu) Aug 4(Fri)
August 3(Thu)

11:00Opening RemarksKen-ichi Kawarabayashi


11:05Jun SakumaUniversity of Tsukuba
Differentially Private Chi-squared Test by Unit Circle MechanismICML'17


11:25Tomoya SakaiThe University of Tokyo
Semi-Supervised Classification Based on Classification from Positive and Unlabeled DataICML'17


11:45Weihua HuThe University of Tokyo
Learning Discrete Representations via Information Maximizing Self-Augmented TrainingICML'17


12:05Emtiyaz KhanRIKEN
Conjugate-Computation Variational Inference: Converting Variational Inference in Non-Conjugate Models to Inferences in Conjugate ModelsAISTATS'17


12:25 - 12:30Questions and Answers


12:30 - 14:00Lunch


14:00Yusuke KobayashiUniversity of Tsukuba
A Weighted Linear Matroid Parity AlgorithmSTOC'17


14:20Kaito FujiiThe University of Tokyo
Budgeted stream-based active learning via adaptive submodular maximizationNIPS'16


14:40Yuichi YoshidaNational Institute of Informatics
Minimizing Quadratic Functions in Constant Time (coauthored with Kohei Hayashi)NIPS'16


15:00Man-Kwun ChiuNational Institute of Informatics
High Dimensional Consistent Digital SegmentsSoCG'17


15:20 - 15:25Questions and Answers


15:25 - 15:40Coffee Break


15:40Naoto OhsakaThe University of Tokyo
Portfolio Optimization for Influence SpreadWWW'17


16:00Wataru InaribaThe University of Tokyo
Random-Radius Ball Method for Estimating Closeness CentralityAAAI'17


16:20Satoshi HaraNational Institute of Informatics
Enumerate Lasso Solutions for Feature SelectionAAAI'17


16:40Tasuku SomaThe University of Tokyo
Regret Ratio Minimization in Multi-objective Submodular Function MaximizationAAAI'17


17:00 - 17:05Questions and Answers


17:05 – 17:20Coffee Break


17:20Shinji ItoNEC Data Science Research Laboratories
Large-Scale Price Optimization via Network FlowNIPS'16


17:40Akihiro YabeNEC Data Science Research Laboratories
Robust Quadratic Programming for Price OptimizationIJCAI'17


18:00Masakazu IshihataHokkaido University
Statistical Emerging Pattern Mining with Multiple Testing CorrectionKDD'17


18:20Naoki MarumoNTT Communication Science Laboratories
SVD-Based Screening for the Graphical LassoIJCAI'17


18:40 - 18:45Questions and Answers


18:45 – 18:50Discussion


19:00 – 20:45Banquet


August 4(Fri)

10:00Tomoharu IwataNTT Communication Science Laboratories
Multi-view Anomaly Detection via Robust Probabilistic Latent Variable ModelsNIPS'16


10:20Atsutoshi KumagaiNTT Secure Platform Laboratories
Learning Latest Classifiers without Additional Labeled DataIJCAI'17


10:40Sho YokoiTohoku University
Learning Co-Substructures by Kernel Dependence MaximizationIJCAI'17


11:00Yasushi KawaseTokyo Institute of Technology
Near-feasible stable matchings with budget constraintsIJCAI'17


11:20 - 11:25Questions and Answers


11:25 - 11:35Coffee Break


11:35Mahito SugiyamaNational Institute of Informatics
Tensor Balancing on Statistical ManifoldICML'17


11:55Hanna SumitaNational Institute of Informatics
Online Optimization of Video-Ad AllocationIJCAI'17


12:15Masato AbeNational Institute of Informatics
Optimizing mating encounters by sexually dimorphic movementsJournal of the Royal Society Interface


12:35 - 12:40Questions and Answers


12:40 - 14:00Lunch


14:00Makoto YamadaRIKEN
Convex Factorization Machine for Toxicogenomics PredictionKDD'17


14:20Masaaki ImaizumiThe Institute of Statistical Mathematics
Tensor Decomposition with SmoothnessICML'17


14:40Daisuke HatanoNational Institute of Informatics
Computing Least Cores of Supermodular Cooperative GamesAAAI'17


15:00Yuto YamaguchiIndeed
When Does Label Propagation Fail? A View from a Network Generative ModelIJCAI'17


15:20 - 15:25Questions and Answers


15:25 - 15:40Coffee Break


15:40Takanori MaeharaRIKEN
Exact Computation of Influence Spread by Binary Decision DiagramsWWW'17


16:00Yasuhiro FujiwaraNTT Software Innovation Center
Scaling Locally Linear EmbeddingSIGMOD'17


16:20Tomohiro SonobeNational Institute of Informatics
Coarsening Massive Influence Networks for Scalable Diffusion AnalysisSIGMOD'17


16:40Suguru TamakiKyoto University
Beating Brute Force for Systems of Polynomial Equations over Finite FieldsSODA'17


17:00Shuichi HiraharaThe University of Tokyo
On the average-case complexity of MCSP and its variantsCCC'17


17:20 - 17:25Questions and Answers


17:25Crosing RemarksKen-ichi Kawarabayashi


Copyright © Kawarabayashi Large Graph Project