[ English | Japanese ]

News


May 2017
  • Our three papers has been accepted to ICML2017.
Jan. 2017
Dec. 2016
  • 5-10th: At NIPS2016, we present our work on theoretical analysis of learning from positive and unlabeled. Prof. Masashi Sugiyama serves as a General Co-Chair for NIPS2016.
  • 19th: Mr. Shane Gu (Cambridge University, UK and Max Planck Institute for Intelligent Systems, Germany) is visiting us and giving a talk on deep reinforcement learning for robotics.
  • 19th: Prof. Heng Tao Shen (The University of Queensland, Australia) is visiting us and giving a talk on hashing big multimedia data.
  • 27th: Prof. Masashi Sugiyama was selected for the 13th JSPS Prize.
Nov. 2016
  • Our paper on contextual reinforcement learning has been accepted to AAAI2017.
  • 9th: Dr. Parag Rastogi (EPFL, Switzerland) is visiting us and giving a talk on machine learning for building design.
  • 16-18th: At ACML2016, we present our works on multitask principal component analysis and stationary subspace analysis.
  • 16-19th: At IBIS2016, we present our works on robust learning, structural change detection, contextual policy search, and semi-supervised classification. Voot Tangkaratt (doctor course) received the Best Student Presentation Award at IBIS2016.
  • 21st: Prof. Eric Xing (CMU, USA) is visiting us and giving a talk on distributed machine learning.
Oct. 2016
  • Our papers on multitask principal component analysis and stationary subspace analysis have been accepted to ACML2016.
  • Our papers on dimension reduction has been accepted to TAAI2016.
  • Our papers on class-prior estimation and geometry-aware PCA have been accepted to Machine Learning.
  • 6th: Dr. Takayuki Osa (Technical University Darmstadt, Germany) is visiting us and giving a talk on learning robotic grasping.
  • 18th: At ICONIP2016, we present our work on modal regression.
  • 31st: Dr. Mohammad Emtiyaz Khan (RIKEN, Japan) is visiting us and giving a talk on approximate Bayesian inference.
Sep. 2016
  • Futoshi Futami (doctor), Nontawat Charoenphakdee (master), Jiangqi Dong (master), Han Bao (bachelor), Zhenghang Cui (bachelor), Yutaka Iso (bachelor), Yusuke Konno (bachelor), and Takuma Yamashita (bachelor) joined Sugiyama-Sato Lab.
Aug. 2016
  • Our paper on theoretical analysis of learning from positive and unlabeled data has been accepted to NIPS2016.
  • Our paper on reinforcement learning with dimension reduction has been accepted to Neural Networks.
  • 1-4th: At MIRU2016, we have invited presentation on structure learning of Markov networks.
  • 2nd: Dr. Bin Yang (Rakuten Institute of Technology, Japan) is visiting us and giving a talk on differential privacy.
  • 3rd: Matthew Holland (Nara Institute of Science and Technology, Japan) is visiting us and giving a talk on stable learning.
  • 4th: Nathan Srebro (Toyota Technological Institute at Chicago, USA) is visiting us and giving a talk on geometry of multilayer networks.
  • 21-24th: At CASE2016, we present our work on camera-LiDAR calibration.
  • 26th: At Asia-Pacific Econophysics Conference (APEC2016), Prof. Masashi Sugiyama gives a talk on machine learning.
  • 31st: At ZMP Forum, Prof. Masashi Sugiyama gives a talk on machine learning.
Jul. 2016
  • Prof. Masashi Sugiyama became the director of the RIKEN Center for Advanced Integrated Intelligence Research.
  • Our paper on mobile robot localization has been accepted to Advanced Robotics.
  • Our paper on modal regression has been accepted to ICONIP2016.
  • Lab alumni party is held.
  • 17-21st: At IEEE-PES2016, we present our work on diagnosis of cyber attacks.
Jun. 2016
  • Nao Inui (assistant technical staff) joined Sugiyama-Sato Lab.
  • Our paper on camera-LiDAR calibration has been accepted to CASE2016.
  • 2nd: At First Korea-Japan Machine Learning Symposium, Prof. Masashi Sugiyama gives a talk on machine learning from weak supervision.
  • 7th: At the 29th Annual Conference of the Japanese Artificial Intelligence Society, Prof. Masashi Sugiyama introduces the RIKEN Center for Advanced Integrated Intelligence Research.
  • 10th: Stephan Zheng (California Institute of Technology, USA) is visiting us and giving a talk on reinforcement learning.
  • 10th: At JSPS Meeting, Prof. Masashi Sugiyama gives a talk on machine learning from limited information.
  • 19-24th: At ICML2016, we present our work on structure learning of Markov networks.
  • 25-29th: At UAI2016, we present our work on stochastic variational inference.
May 2016
  • Our paper on stochastic variational inference has been accepted to UAI2016.
Apr. 2016
  • Seiya Tokui (doctor course), Soma Yokoi (doctor course), Masayoshi Hayashi (master course), Weihua Hu (master course), Hideaki Kano (master course), Ryosuke Kamesawa (master course), Ryuichi Kiryo (master course), Taishi Watanabe (master course), Nontawat Charoenphakdee (research student), and Etsuko Yoshida (assistant technical staff) joined Sugiyama-Sato Lab.
  • Our paper on theoretical analysis of structural change detection has been accepted to The Annals of Statistics.
  • Our paper on structure learning of Markov networks has been accepted to ICML2016.
  • 6th: Mr. Wittawat Jitkrittum (University College London, UK) is visiting us and giving a talk on statistical tests.
  • 25th: At First Joint Symposium on Next Generation AI Technology, Prof. Masashi Sugiyama introduces the RIKEN AIP Center.
Mar. 2016
Feb. 2016
Jan. 2016
Dec. 2015
  • Our paper on humanoid motor learning has been accepted to IEEE Robotics & Automation Magazine.
  • Our paper on non-Gaussian component analysis has been accepted to AISTATS2016.
  • 7-10th: Prof. Masashi Sugiyama serves as a Program Co-Chair for NIPS2015.
  • 11-12th: At NIPS2015 Workshops, we present our works on proximal-gradient stochastic variational inference and metric learning for non-parametric regression.
  • 16th: Mr. Kangasrääsiö Antti (Aalto University, Finland) is visiting us and giving a talk on exploratory search.
Nov. 2015
  • Our paper on tensor regression and classification has been accepted to Neural Computation.
  • Our paper on diagnosis of cyber attacks has been accepted to IEEE-PES2016.
  • 5-6th: At Frontiers in Data-Driven Science and Technology Workshop, Prof. Masashi Sugiyama gives a talk on recent advances in learning from small data.
  • 16th: Prof. Samuel Kaski (Aalto University, Finland) is visiting us and giving a talk on Bayesian factorization.
  • 18th: Dr. Shinichi Nakajima (Technical University of Berlin, Germany) is visiting us and giving a talk on structured learning.
  • 20-22nd: At ACML2015, we present our works on geometry-aware PCA, class-prior estimation, target shift adaptation, regularized policy gradients, and sufficient dimension reduction.
  • 20-22nd: At TAAI2015, we present our work on crowdsourcing.
  • 24th: At Science Council of Japan Workshop, Prof. Masashi Sugiyama gives a talk on learning from small data and Lect. Issei Sato gives a talk on Bayesian optimization.
Oct. 2015
Sep. 2015
  • Issei Sato (lecturer), Gang Niu (project assistant professor), and Atsuko Sano (assistant technical staff) joined Sugiyama-Sato Lab.
  • Masayoshi Hayashi (bachelor course), Ryosuke Kamesawa (bachelor course), Ryuichi Kiryo (bachelor course), Yurina Ono (bachelor course), and Yoshiki Yamaguchi (bachelor course) joined Sugiyama-Sato Lab.
  • Kiyoshi Irie (doctor course) was selected for the RSJ Incentive Award.
  • Our paper on covariance estimation has been accepted to Pattern Recognition Letters.
  • Our papers on target shift adaptation, regularized policy gradients, and sufficient dimension reduction have been accepted to ACML2015.
  • Our paper on task selection in crowdsourcing has been accepted to TAAI2015.
  • Our paper on class prior estimation has been accepted to IEICE Transactions.
  • 7th: Mr. Daniele Calandriello (INRIA, France) is visiting us and giving a talk on semi-supervised learning.
  • 7-11th: At ECMLPKDD2015, we present our work on sparse conditional density estimation.
  • 24th: At Toyota Technological Institute, Prof. Masashi Sugiyama gives a talk on machine learning.
  • 28th-Oct. 2nd: At IROS2015, we present our work on mobile robot localization.
Aug. 2015
Jul. 2015
Jun. 2015
  • Our papers on geometry-aware PCA and class-prior estimation have been accepted to ACML2015.
  • Our paper on mobile robot localization has been accepted to IROS2015.
May 2015
  • Our paper on medical analysis in developing countries has been accepted to KDD2015.
  • 9-12th: At AISTATS2015, we present our work on density-derivative estimation.
  • 21st: At PAKDD2015, Prof. Masashi Sugiyama gives a keynote speak on change detection.
  • 26th: Prof. Christos Dimitrakakis (Chalmers University of Technology) is visiting us and giving a talk on Bayesian inference and privacy.
Apr. 2015
Mar. 2015
Feb. 2015
  • 6th: Thesis presentation.
  • 10th: Lab party.
Jan. 2015
  • Our paper on density-derivative estimation has been accepted to AISTATS2015.
  • Tomoya Sakai (master course) was selected as JSPS Research Fellow.
  • Our paper on density ratio estimation with convolutional neural networks has been accepted to IEICE Transactions.
  • 1st: Happy new year!
  • 22nd: At Meiji Institute for Advanced Study of Mathematical Sciences, Prof. Masashi Sugiyama gives a talk on density ratio estimation.
  • 29th: At PRESTO meeting, Prof. Masashi Sugiyama gives a talk on reinforcement learning.
Dec. 2014
  • Our paper on cross domain matching has been accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence.
  • 8-11th: At NIPS2014, we present our works on learning from positive and unlabeled data, tensor-based multitask learning, and variational Bayesian latent Dirichlet allocation.
  • 18th: Dr. Makoto Yamada (Yahoo Labs) is visiting us and giving a talk on feature selection.
  • 25th: At IIAS Network Science Meeting, Prof. Masashi Sugiyama gives a talk on density ratio estimation.
  • 25th: At Institute of Statistical Mathematics, Song Liu (postdoctoral fellow) gives a talk on change detection.
Nov. 2014
  • Our paper on theoretical analysis of structural change detection has been accepted to AAAI2015.
  • 14th: Prof. Aapo Hyvarinen (University of Helsinki) is visiting us and giving a talk on dynamic connectivity factorization.
  • 17-19th: At IBIS2014, we present our works on machine learning. Hiroaki Sasaki was selected for the Best Young Researcher Award, and Hyunha Nam was selected for the Best Student Award.
  • 18-20th: At HUMANOIDS2014, we present our work on humanoid robot control.
  • 21st: At JSAI-SIG Challenge Meeting, Prof. Masashi Sugiyama gives a talk on density ratio estimation.
Oct. 2014
  • Masashi Sugiyama became a professor at The University of Tokyo.
  • Yuko Kawashima (secretary) joined Sugiyama Lab.
  • Our paper on conditional density estimation has been accepted to Machine Learning.
  • 20th: Mr. Julian Zimmert (Humboldt University of Berlin) is visiting us and giving a talk on multi-task learning.
Sep. 2014
  • Our paper on image registration has been accepted to Precision Engineering.
  • Our papers on learning from positive and unlabeled data, tensor-based multitask learning, and variational Bayesian latent Dirichlet allocation have been accepted to NIPS2014.
  • Our paper on humanoid robot control has been accepted to HUMANOIDS2014.
  • Our project proposal "Understanding, Summarization, and Retrieval of Massive Multimedia Data (Leader: Prof. Tasuya Harada, University of Tokyo)" has been accepted for CREST.
  • 15-19th: At ECMLPKDD2014, we present our works on mean-shift clustering and online Markov decision problems.
Aug. 2014
Jul. 2014
Jun. 2014
  • Our papers on mean-shift clustering and online Markov decision problems have been accepted to ECMLPKDD2014.
  • Our paper on model-based reinforcement learning has been accepted to Neural Networks.
  • 16th: At MLSS2014, we give a lecture on unsupervised change detection.
  • 21-26th: At ICML2014, we present our works on multi-class transduction and robust classification.
  • 30th: Mr. Chiyuan Zhang (MIT) is staying with us for a month. He is giving a talk on data representation.
May 2014
Apr. 2014
  • Hiroaki Sasaki (postdoctoral fellow), Marthinus Christoffel du Plessis (postdoctoral fellow), Song Liu (postdoctoral fellow), Voot Tangkaratt (doctor course), Hiroaki Shiino (master course), and Inbal Horev (Japanese trainee) joined Sugiyama Lab.
  • Our paper on density difference estimation has been accepted to IEICE Transactions.
  • Our papers on multi-class transduction and robust classification have been accepted to ICML2014.
  • 11th: Prof. Tony Jebara (Columnbia University) is visiting us and giving a talk on quadratic majorization.
  • 15th: Prof. Masashi Sugiyama was selected for the Young Scientists' Prize from the Commendation for Science and Technology by the Minister of Education, Culture, Sports, Science and Technology.
  • 22nd-24th: At AISTATS2014, we present our works on KL-divergence estimation and theoretical analysis of variational Bayes. The paper on theoretical analysis of variational Bayes was selected for the Notable Paper Award.
Mar. 2014
  • Alessandro Balzi (Politecnico di Milano) is staying with us for half a year.
  • Our paper on multi-task learning has been accepted to IEICE Transactions.
  • 18th: At Japan Society for Precision Engineering 2014 Student Meeting, we present our work on image registration.
  • 19th: At Nonparametric Statistical Analysis and Bayesian Statistics, we give a talk on density ratio estimation.
Feb. 2014
  • Florian Yger (postdoc) joined Sugiyama Lab.
  • Our paper on statistical analysis of distance estimation has been accepted to Entropy.
  • Our paper on class-prior estimation has been accepted to IEICE Transactions.
  • Our paper on semi-supervised metric learning has been accepted to Neural Computation.
  • 24th: At Kinki Chemical Society, we give a talk on density ratio estimation.
  • 28th: Dr. Alessio Del Bue (Italian Institute of Technology) is visiting us and giving a talk on Bilinear modeling with missing data.
Jan. 2014
  • 1st: Happy new year!
  • Our paper on structural change detection has been accepted to Neural Computation.
  • The PhD dissertation by Ning Xie was selected for Tejima Doctor Dissertation Award.
  • Our papers on KL-divergence estimation and theoretical analysis of variational Bayes have been accepted to AISTATS2014.
Dec. 2013
  • Our paper on squared-loss mutual information estimation has been accepted to IEICE Transactions.
  • Our paper on anomaly detection has been accepted to Pattern Recognition Letters.
  • 3rd: Prof. Aapo Hyvarinen (University of Helsinki) is visiting us and giving a talk on independent component analysis.
  • 5-8th: At NIPS2013, we present our works on multi-task feature learning and variational Bayesian subspace clustering.
  • 6-8th: At TAAI2013, we present our work on unsupervised labeling. This paper was selected for Best Paper Award.
  • 9th: At NIPS2013 Workshop, we present our work on multi-task learning.
  • 21st: At NC, we give a talk on conditional density estimation.
Nov. 2013
  • Our paper on class-prior estimation has been accepted to Neural Networks.
  • 3rd: At ICONIP2013, we give a tutorial on density ratio estimation.
  • 5th: At KAIST, we give a talk on change detection.
  • 10-13th: At IBIS2013, we present our works on support vector comparison, variational Bayesian subspace clustering, information estimation, feature selection in conditional density estimation, feature extraction in conditional density estimation, robust support vector machine training, parametric task learning, sumi-e stroke rendering by reinforcement learning, online policy gradient method, online density ratio estimation, road recognition, class prior estimation, budget allocation in crowdsourcing, and log-density derivative estimation.
  • 18th: Prof. Taiji Suzuki (Tokyo Institute of Technology) is visiting us and giving a talk on convex tensor decomposition.
  • 20th: Dr. Shinichi Nakajima (Nikon Corporation) is visiting us and giving a talk on variational Bayesian low-rank subspace clustering.
  • 20th: Prof. Kazuho Watanabe (Nara Institute of Science and Technology) is visiting us and giving a talk on rate-distortion theory.
  • 27th: Mr. Hiroshi Kajino (University of Tokyo) is visiting us and giving a talk on crowdsourcing.
  • 28-29th: At IPSJ-GCAD, we preesnt our work on sumi-e stroke rendering by reinforcement learning.
  • 29th: At Iron and Steel Institute of Japan, we give a talk on change detection.
  • 30th: At WiNF2013, we present our work on robust support vector machines. This paper was selected for Best Paper Award.
Oct. 2013
  • Gang Niu (postdoc), Kiyoshi Irie (doctor course), and Hideko Kawakubo (doctor course) joined Sugiyama Lab.
  • Prof. Yung-Kyun Noh (KAIST) is staying with us for a month.
  • Mr. Pantelis Pipergias Analytis (Max Planck Institute for Human Development) is staying with us for 4 months.
  • Mr. Erling Wishman Eeg-Henriksen (Norwegian University of Science and Technology) is staying with us for a year.
  • Our paper on causal inference has been accepted to Machine Learning.
  • We start a lecture series on machine learning.
  • 8th: Dr. Hiroaki Sasaki (University of Electro-Communications, Japan) is visiting us and giving a talk on topograpy estimation.
  • 15th: At Tsinghua University, we give a talk on density ratio estimation.
  • 16th: At Peking University, we give a talk on change detection.
  • 29th: At MATLAB Expo 2013, we give a talk on machine learning.
Sep. 2013
  • Ms. Atik Wintarti Imam Supartono (Institut Teknologi Sepuluh Nopember Surabaya) is staying with us for 3 months.
  • Our papers on multi-task feature learning and variational Bayesian subspace clustering have been accepted to NIPS2013.
  • Our paper on unsupervised labeling has been accepted to TAAI2013.
  • 2nd: At IBISML, we give a talk on a training algorithm for robust support vector machine.
  • 26th: At University of Toronto, we give a talk on density ratio estimation.
  • 27th: At Toyota Technological Institute Chicago, Tenth Anniversary Symposium, we give a talk on density ratio estimation.
Aug. 2013
  • Our survey paper on learning under non-stationarity has been accepted to WIREs Computational Statistics.
  • Our paper on information-maximization clustering has been accepted to Neural Computation.
  • Our paper on high-dimensional feature selection has been accepted to Neural Computation.
  • 2nd: Thesis presentation/evaluation --> Lab party.
  • 5-6th: At MLAB2013, we present our works on change detection and graph layout comparison.
  • 9th: Prof. Taku Komura (University of Edinburgh) is visiting us and giving a talk on generating animation.
Jul. 2013
  • Our paper on quadratic mutual information clustering has been accepted to IEICE Transactions.
  • Our paper on maximum volume clustering has been accepted to JMLR.
  • 3rd: Dr. Shinichi Nakajima (Nikon Corporation) is visiting us and giving a talk on variational Bayesian PCA.
  • 18th: At IBISML, we give talks on unsupervised labeling, quadratic mutual information clustering, and structural change detection.
  • 22nd: Prof. Daniel Lee (University of Pennsylvania) is visiting us and giving a talk on perception, planning, and motor control.
Jun. 2013
  • Our paper on structural change detection has been accepted to ECMLPKDD2013.
  • 6th: Prof. Taiji Suzuki (University of Tokyo) is visiting us and giving a talk on stochastic optimization.
  • 13th: At SSII2013, we give a talk on divergence estimation.
  • 16-21st: At ICML2013, we present our works on information-theroretic semi-supervised learning and semi-supervised support vector machine.
  • 26th: We organize a seminar on Machine Learning for Big Data.
May 2013
Apr. 2013
  • Duong Tuan Nguyen (master course), Tomoya Sakai (master course), Hiroyuki Tanaka (master course), Ikko Yamane (master course), Hiroaki Shiino (bachelor course), Dang Hai Tran (bachelor course) and, Hideko Kawakubo (assistant) joined Sugiyama Lab.
  • Our project on information-based machine learning was accepted to Grant-in-Aid for Young Scientists (A).
  • Our paper on variational Bayesian matrix factorization has been accepted to Machine Learning.
  • Our paper on information-theoretic feature selection by sparse learning has been accepted to IEICE Transactions.
  • Our paper on density-difference estimation has been accepted to Neural Computation.
  • Our papers on information-theroretic semi-supervised learning and semi-supervised support vector machine have been accepted to ICML2013.
  • Our paper on multi-label probabilistic classification has been accepted to IEICE Transactions.
  • 3rd: Prof. Takayuki Okatani (Tohoku University) is visiting us and giving a talk on deep learning.
  • 4th: Prof. Ichiro Takeuchi (Nagoya Institute of Technology) is visiting us and giving a talk on multi-task learning.
  • 11th: Prof. Masahiro Tomono (Chiba Institute of Technology) is visiting us and giving a talk on environment sensing for mobile robots.
  • 12th: We have a welcome party.
  • 24th: At Kyoto University, we give a tutorial on machine learning.
  • 25th: At Omron, we give a talk on density ratio estimation.
Mar. 2013
Feb. 2013
  • Our paper on application of reinforcement learning to sumie painting has been accepted to IEICE Transactions.
  • 7th: Master thesis presentation.
  • 15th: Bachelor thesis presentation --> Lab party.
  • 19th: At BCI2013, we give a talk on divergence estimation.
  • 22nd: At Seoul National University, we give a talk on density ratio estimation.
  • 26th: At NTT CS Lab, we give a talk on divergence estimation.
Jan. 2013
  • Toby Dylan Hocking (postdoc) and Ning Xie (postdoc) joined Sugiyama Lab.
  • Our paper on a policy gradient method in reinforcement learning has been accepted to Neural Computation.
  • Our paper on change detection has been accepted to Neural Networks.
  • Our paper on multiple kernel learning has been accepted to The Annals of Statistics.
  • 1st: Happy new year!
  • 24th: Prof. Makoto Okabe (University of Electro-Communications, Tokyo) is visiting us and giving a talk on video retrieval and animation synthesis.
Dec. 2012
  • Our paper on relative density-ratio estimation has been accepted to Neural Computation.
  • Our paper on macine learning with squared-loss mutual information has been accepted to Entropy.
  • 3-6th: At NIPS2012, we present our works on density-difference estimation and variational Bayesian PCA.
  • 8th: At OPT2012, we present our work on the homotopy method.
  • 14th: At PRESTO Symposium, we give a talk on density ratio estimation.
Nov. 2012
Oct. 2012
  • Yao Ma (doctor course), Takatoshi Yoshida (bachelor course), Hao Zhang (research student), and Daniele Calandriello (YSEP exchange student) joined Sugiyama Lab.
  • Song Liu (doctor course) was selected as a JSPS Research Fellow.
  • 6-7th: Open lab. at Koudaisai.
  • 7th: Joint alumni party of Iijima-Ogawa-Sugiyama Lab. is held.
Sep. 2012
  • Our paper on active learning has been accepted to Neural Networks.
  • Our papers on density-difference estimation and variational Bayesian PCA have been accepted to NIPS2012.
  • Our paper on sparse additive matrix factorization has been accepted to ACML2012.
  • Our paper on numerical analysis of density ratio estimation has been accepted to Machine Learning.
  • 7th: At MLSS2012, we give a lecture on density-ratio estimation.
  • 19-20th: At IPSJ SIG MPS, we present our works on Salient object detection, event detection, multivariate component analysis, and semi-supervised canonical correlation analysis.
  • 25th: At BBCI2012, we give a lecture on density-ratio estimation.
  • 28-29th: At SICE Plant-Modeling Meeting, we give a lecture on machine learning.
Aug. 2012
  • Our paper on frequency-domain independent component analysis has been accepted to Digital Signal Processing.
  • Our paper on change detection has been accepted to SPR2012.
  • Our paper on optical surface profiling has been accepted to Applied Optics.
  • 6-7th: At MLAB2012, we present our work on divergence estimation.
  • 6-8th: At MIRU2012, we present our works on semi-supervised media annotation and foreground-background video separation.
  • 8th: At Hokkaido University, we give a talk on estimation of L2-distance between probability densities.
  • 30th: At JSIAM2012, we organize a session on machine learning.
Jul. 2012
  • Ayako Tamai (secretary) joined Sugiyama Lab.
  • Our paper on a multi-task reinforcement learning has been accepted to IEICE Transactions.
  • 4th: At IMS-APRM2012, we organize a session on density ratio estimation.
  • 6th: Mr. Mathieu Blondel (Kobe University) is visiting us and giving a talk on sparse learning.
Jun. 2012
  • Our paper on sufficient dimension reduction has been accepted to Neural Computation.
  • Our paper on generalized component analysis has been accepted to ICPR2012.
  • Our paper on a kernel-based independence measure has been accepted to IEICE Transactions.
  • Our paper on canonical dependency analysis has been accepted to Neural Networks.
  • Our paper on relative density-ratio estimation was selected for IBISML Award Finalist.
  • 11th: At Keio University, we give a talk on density ratio estimation.
  • 18th: At Decisions area meeting, we give a talk on feature selection.
  • 19-20th: At IBISML, we present our works on density-difference estimation, Kaggle challenge, and semi-supervised support vector machines.
May 2012
  • Our papers on semi-supervised metric learning, oriental ink painting by reinforcement learning, and semi-supervised learning under class-balance change have been accepted to ICML2012.
Apr. 2012
  • Ildefons Magrans de Abril (postdoc), Janya Sainui (doctor course), Voot Tangkaratt (master course), Norimi Ishii (master course), Akihiro Yamashita (master course), Satoshi Yoshida (master course), Duong Tuan Nguyen (bachelor course), Tomoya Sakai (bachelor course), and Keisuke Nakata (bachelor course) joined Sugiyama Lab.
  • Our paper on incremental active learning for probabilistic classification has been accepted to IEICE Transactions.
  • 12th: Prof. Masashi Sugiyama attends the panal discussion at SICE Plant Modeling Symposium on "What is a Model?".
  • 13th: At SICE Plant Modeling Symposium, we give a tutorial on pattern recognition and machine learning.
  • 20th: Welcome party.
  • 21-23rd: At AISTATS2012, we present our work on multiple kernel learning.
  • 25th: At the 3rd Symposium on Computational Science Simulation, we give a talk on density ratio estimation.
Mar. 2012
  • Prof. Masashi Sugiyama was selected for Funai Award (Funai Foundation for Information Technology).
  • Our book on density ratio estimation was published (Cambridge University Press).
  • Our book on covariate shift adaptation was published (MIT Press).
  • Our paper on weighted support vector machines has been accepted to Machine Learning.
  • 5th: Dr. Minh Ha Quang (Italian Institute of Technology) is visiting us and giving a talk on Vector-valued RKHS and slow feature analysis.
  • 9th: At Decisions area meeting, we give a talk on reinforcement learning.
  • 12-13th: At IBISML, we present our works on class-prior estimation, semi-supervised learning, incremental active learning, active learning, feature selection, and reinforcement learning.
  • 14-16th: At The Japan Society for Precision Engineering, 2012 Spring Semestrial Conference, we present our work on one-shot surface profiling.
  • 26th: Wittawat Jitkrittum, Tsubasa Kobayashi, Nozomi Kurihara, and Jun Takagi received the M.E. degree, and Yuichi Sayama and Akihiro Yamashita received the B.E. degree.
  • 28th: At ICASSP2012, we give a poster presentation on multi-label classification.
  • 30th: Prof. Hal Daume III (University of Maryland) is visiting us and giving a talk on structured prediction.
  • 30th: Prof. Ruslan Salakhutdinov (University of Toronto) is visiting us and giving a talk on learning hierarchical models.
  • 31st: At IWSML2012, we present our work on probabilistic classification.
Feb. 2012
  • 7th: Master thesis presentation.
  • 14th: Bachelor thesis presentation --> Lab party.
  • 17th: At IPAB, we give a talk on probabilistic pattern recognition and clustering.
Jan. 2012
Dec. 2011
  • Our paper on theoretical analysis of multiple kernel learning has been accepted to AISTATS2012.
  • Our paper on multi-label classification has been accepted to ICASSP2012.
  • 2nd: At Data Science Robotics Meeting, we give a tutorial on regression and classification.
  • 8-9th: At ViEW2011, we present our work on event detection from videos.
  • 10th: At Empirical Inference Symposium, we present our work on density ratio estimation.
  • 13-15th: At NIPS2011, we present our work on variational Bayesian matrix factorization, theoretical analysis of policy gradient estimation, feature selection for nearest neighbor classifiers, and relative density ratio estimation.
Nov. 2011
  • 2nd: Dr. Moritz Grosse-Wentrup (Max-Planck Institute) is visiting us and giving a talk on brain-computer interface.
  • 4th: Prof. Marco Cuturi (Kyoto University) is visiting us and giving a talk on metric learning.
  • 5th: Joint alumni party of Iijima-Ogawa-Sugiyama Lab. is held.
  • 8th: At Decisions group meeting, we give a talk on feature selection.
  • 9-11th: At IBIS2011, we present our works on relative density-ratio estimation, stable policy search, generalized matrix factorization, change detection, and multi-label classification.
  • 13th: At VECTaR2011, we present our work on event detection from videos.
  • 13-15th: At ACML2011, we present our work on sufficient dimension reduction.
  • 16th: At National Cheng Kung University, we give talks on density ratio estimation.
  • 17th: At National Taiwan University, we give talks on density ratio estimation.
  • 22nd: At E-Science Symposium, we give a talk on density ratio estimation.
  • 28-30th: At ACPR2011, we present our work on age prediction from face images.
Oct. 2011
Sep. 2011
  • Rubaiya Khan received the B.E. degree and Liao Wei-Pei finished the YSEP course.
  • Our paper on sufficient dimension reduction has been accepted to ACML2011.
  • Our paper on human activity recognition has been accepted to Neurocomputing.
  • Our paper on policy gradient estimation in reinforcement learning has been accepted to Neural Networks.
  • Our paper on statistical analysis of kernel-based density-ratio estimation has been accepted to Machine Learning.
  • 5-6th: At IBISML, we present our works on variational Bayesian matrix factorization, application of reinforcement learning to computer graphics, and canonical dependency analysis.
  • 6-8th: At JIASC2011, we present our work on regions-of-interest detection in images.
  • 15th: At Hokkaido University, we give a talk on probabilistic pattern recognition and clustering.
  • 18-21st: At MLSP2011, we present our work on regularization path tracking.
  • 30th: We have a 5th-floor joint party.
Aug. 2011
  • Our paper on theoretical analysis of variational Bayesian matrix factorization has been accepted to JMLR.
  • Our paper on age prediction from face images has been accepted to ACPR2011.
  • Our paper on event detection from videos has been accepted to VECTaR2011.
  • Our papers on variational Bayesian matrix factorization, theoretical analysis of policy gradient estimation, feature selection for nearest neighbor classifiers, and relative density ratio estimation have been accepted to NIPS2011.
  • Our research achievement on optical surface profiling was selected for JSPE Technology Award.
  • 4th: ICML2011 seminar is held.
  • 5th: Mr. Oliver Kroemer (Max Planck Institute for Intelligent Systems, Germany) is visiting us and giving a talk on learning tactile sensing.
  • 7-11th: At AAAI2011, we present our works on hetero-distributional subspace analysis and trajectory regression.
  • 12th: At Yahoo! (Calfornia), we give talks on density ratio estimation.
  • 23rd: At ERATO Minato Discrete Structure Manipulation Sytem Project (Hokkaido University), we give a talk on density ratio estimation.
Jul. 2011
  • Our project on "Elucidation of neural computation for prediction and decision making: toward better human understanding and applications" (leader: Prof. Kenji Doya) was accepted to Grant-in-Aid for Scientific Research on Innovative Areas.
  • Our paper on semiparametric two-sample test has been accepted to IEEE Transactions on Information Theory.
  • Our paper on regularization path tracking has been accepted to MLSP2011.
  • 1st: At ICML2011, we present our works on information-maximization clustering, theoretical analysis of Bayesian PCA, and approximate regularization-path tracking.
  • 5th: At NEC Labs America (Calfornia), we give talks on density ratio estimation and regularization path tracking.
Jun. 2011
  • 16th: Prof. Pritee Khanna (Indian Institute of Information Technology) is visiting us and giving a talk on content-based image retrieval.
  • 20-21st: At IBISML, we present our works on mutual information estimation, semi-supervised metric learning, variance analysis of poilcy search in reinforcment learning, and direct density-ratio estimation with dimensionality reduction.
May 2011
Apr. 2011
  • Masayuki Karasuyama (postdoc), Song Liu (doctor course), HyunHa Nam (doctor course), Christo du Plessis (doctor course), Tingting Zhao (doctor course), Tomoaki Oiki (master course), Syogo Mori (master course), Yuichi Sayama (bachelor course), and Akihiro Yamashita (bachelor course) joined Sugiyama Lab.
  • Our paper on optimization techniques for sparse regularized estimation has been accepted to JMLR.
  • Our paper on change detection has been accepted to Statistical Analysis and Data Mining.
  • Our paper on two-sample test has been accepted to Neural Networks.
  • Our papers on hetero-distributional subspace analysis and trajectory regression have been accepted to AAAI2011.
  • Our papers on information-maximization clustering, Bayesian PCA, and regularization-path tracking have been accepted to ICML2011.
  • Our paper on dependence-maximization clustering has been accepted to JACIII.
Mar. 2011
  • Our papers on cross-domain object matching and maximum volume clustering have been accepted to AISTATS2011.
  • Our paper on theoretical analysis of boosting has been accepted to JMLR.
  • Prof. Masashi Sugiyama was selected for IPSJ Nagao Special Researcher Award.
  • 3-4th: At DIA2011, we present our work on multi-race age classification.
  • 9th: Prof. Jen-Tzung Chien (National Cheng Kung University) is visiting us and giving a talk on Bayesian and Sparse Learning of Acoustic and Language Models.
  • 17th: Mr. Masanao Ochi (University of Electro-Communications) is visiting us and giving a talk on features for ranking.
  • 26th: Yuta Ito and Tatsuya Shigemura received the M.E. degree, and Taketo Akama, Takehiko Mizoguchi, and Syogo Mori received the B.E. degree.
  • 28-29th: At IBISML, we present our works on information-maximization clustering, Bayesian PCA, statistical properties of kernel density-ratio estimators, and theoretical analysis of multiple kernel learning.
  • 29-30th: At DMSS2011, we present our works on maximum volume clustering, sufficient dimension reduction, dependence-maximization clustering, feature weighting for nearest neighbor classifiers, and acceloerometer-based human activity recognition.
Feb. 2011
Jan. 2011
Dec. 2010
  • 1st: At Fujitsu Laboratories in Kawasaki, we give a talk on density ratio estimation.
  • 6-9th: At NIPS2010, we present our work on variational Bayesian matrix factorization.
  • 9-10th: At IEICE PRMU, we present our work on audio tagging-retrieval.
  • 9-10th: At ViEW2010, we present our work on region-of-interest detection in images.
  • 11th: At NIPS2010 Workshop on Transfer Learning Via Rich Generative Models, we present our work on accelerometer-based human activity recognition.
  • 22nd: At Toshiba in Kawasaki, we give a talk on conditional probability estimation.
Nov. 2010
  • Our paper on age prediction from face images has been accepted to IEICE Transactions.
  • Makoto Yamada (postdoc) received the honorable mention for his presentation at IBIS2010.
  • Our paper on multi-task learning has been accepted to IPSJ Transaction.
  • 4-6th: At IBIS2010, we present our works on density ratio estimation, variational Bayesian matrix factorization, cross-domain object matching, return density estimation in reinforcement learning, and audio tagging-retrieval.
  • 8-10th: At ACML2010, we present our works on fast nearest neighbor search, feature selection in reinforcement learning, causal inference, model-induced regularization, and maximum volume clustering.
  • 11th: Prof. Thomas G. Dietterich (Oregon State University) is visiting us and giving a talk on visual categorization.
  • 15th: Dr. Fernando Villavicencio (Yamaha Corp.) is visiting us and giving a talk on application of voice-conversion to singing-voice.
  • 20th: Joint alumni party of Iijima-Ogawa-Sugiyama Labs is held.
  • 24th: At Kyoto University, we give a talk on density ratio estimation.
  • 25th: At women researcher promotion program joint symposium, we give a talk on non-stationarity adaptation learning.
  • 30th: At CompView GCOE Symposium, we give a talk on density ratio estimation.
Oct. 2010
  • Gang Niu (doctor course), Christo du Plessis (research student), Tingting Zhao (research studnet), and Liao Wei-Pei (YSEP) joined Sugiyama Lab.
  • Our paper on density ratio estimation in high-dimensional space has been accepted to Neural Networks.
  • 4th: At Google in Tokyo, we give a talk on density ratio estimation.
  • 7th: Dr. Tsuyoshi Ide (IBM Research - Tokyo) is visiting us and giving a talk on trajectory regression.
  • 12th: At Sony in Tokyo, we give a talk on density ratio estimation.
  • 23-24th: Open lab at the campus festival.
  • 26th: Dr. Yasuo Tabei (JST) is visiting us and giving a talk on fast nearest neighbor search.
Sep. 2010
Aug. 2010
  • 23-26th: At ICPR2010, we present our works on age estimation from face images and semi-supervised canonical correlation analysis.
  • 29th-Sep. 1st: At MLSP2010, we present our works on probabilistic classification and support vector regression.
Jul. 2010
  • Our paper on optical surface profiling has been accepted to Applied Optics.
  • Our paper on density ratio estimation has been accepted to IEICE Transactions.
  • Our poster on semi-supervised image recognition was selected for the best interactive session award at MIRU2010.
  • 8-11th: At UAI2010, we present our work on return density estimation in reinforcement learning.
  • 12th: Paper submission deadline of ACML2010.
  • 11-15th: At AAAI2010, we present our work on causal inference.
  • 15th: At Georgia Institute of Technology, we give a talk on density ratio estimation and speech signal processing.
  • 26th: Prof. Ivor Wai-Hung Tsang (Nanyang Technological University, Singapore) is visiting us and giving a talk on multiple kernel learning.
  • 27-29th: At MIRU2010, we present our works on multi-task learning, object recognition, and region-of-interest detection.
Jun. 2010
  • Our research on active learning for reinforcement learning (presented at DMSS2009) was selected for the JSAI Incentive Award.
  • Our paper on return density estimation in reinforcement learning has been accepted to UAI2010.
  • Our paper on independent component analysis has been accepted to Neural Computation.
  • Our paper on robust reinforcement learning has been accepted to IEICE Transactions.
  • Our paper on feature selection for reinforcement learning has been accepted to ECMLPKDD2010.
  • Our paper on age prediction from face images has been accepted to IEICE Transactions.
  • 7th: We give a tutorial on dimensionality reduction at the Institute of Systems, Control and Information Engineers.
  • 11st: Mr. Yukikazu Hidaka (University of Southern California) is visiting us and giving a talk on computational neuro-rehabilitation.
  • 14-15th: At IEICE IBISML, we present our works on density ratio estimation, reinforcement learning, and causal inference.
  • 18-19th: At IPSJ SIG-BIO, we present our work on semi-supervised biological network estimation.
  • 21-24th: At ICML2010, we present our works on Bayesian matrix factorization, large-scale matrix factorization, and risk-sensitive reinforcement learning.
May 2010
  • Our research on probabilistic classification (presented at IEICE PRMU, Jan. 2010) was selected for the PRMU Incentive Award.
  • Our papers on probabilistic classification and support vector regression have been accepted to MLSP2010.
  • 13th-16th: At AISTATS2010, we present our works on conditional density estimation and sufficient dimension reduction.
  • 17th-21st: At VISAPP2010, we present our work on age prediction.
  • 25th: At SICE seminar, we give a lecture on density ratio estimation for machine learning
  • 31st: Dr. Sungyoung Kim (Yamaha Corporation) is visiting us and giving a talk on spatial audio processing.
Apr. 2010
  • Makoto Yamada (postdoc), Manabu Kimura (postdoc), Wittawat Jitkrittum (master course), Tsubasa Kobayashi (master course), and Jun Takagi (master course), Nozomi Kurihara (master course), Taketo Akama (bachelor course), Takehiko Mizoguchi (bachelor course), Syogo Mori (bachelor course) joined Sugiyama Lab.
  • Our review article on optimization algorithms for sparse learning and multiple kernel learning was published in Image Lab.
  • Our papers on age prediction from faces and semi-supervised CCA have been accepted to ICPR2010.
  • Our papers on low-rank matrix approximation, variational Bayesian matrix factorization, and risk-sensitive reinforcement learning have been accepted to ICML2010.
  • 26th: At Carnegie Mellon University, we give a talk on density ratio estimation.
  • 29th-May 1st: At SDM2010, we present our work on high-dimensional density-ratio estimation.
Mar. 2010
  • Hirotaka Hachiya (postdoc) received the Young Researcher Award from IEEE Computational Intelligence Society Japan Chapter.
  • A review paper on machine learning based on density ratio has been accepted to Proceedings of the ISM.
  • Our paper on causality learning has been accepted to AAAI2010.
  • Web page of ACML2010 was open.
  • 1st (to 19th): Ms. Katja Hansen (Technical University of Berlin, Berlin, Germany) is staying with us. She gives a talk on computational chemistry on Mar. 1st.
  • 4th: At IPSJ SIG MPS, we present our work on multi-task learning.
  • 5th: At DIA2010, we present our work on optical surface profiling.
  • 8th: At Research Institute for Mathematical Sciences (Kyoto University), we give a talk on density ratio estimation.
  • 9th: At ASJ 2010 Spring Meeting, we present our work on audio tagging.
  • 14th-19th: At ICASSP2010, we present our work on density ratio estimation, speaker identification, and audio tagging.
  • 23rd: At DAGStat2010, we present our work on partial least squares.
  • 26th: Takayuki Akiyama and Satoshi Hara received the M.E. degree, and Kazuma Inayoshi, Akira Ohgawara, Tsubasa Kobayashi, and Nozomi Kurihara received the B.E. degree.
  • 29th-30th: At JSAI DMSM2010-3, we present our work on density ratio estimation.
Feb. 2010
  • Our paper on thesaurus construction has been accepted to IEICE Transactions.
  • Our papers on conditional density estimation and sufficient dimension reduction has been accepted to AISTATS2010.
Jan. 2010
  • Happy new year!
  • Our paper on theoretical analysis of density ratio estimation has been accepted to IEICE Transactions.
  • Our paper on age estimation has been accepted to VISAPP2010.
  • 21st: We present our work on probabilistic classification at IEICE PRMU.
Dec. 2009
  • Our paper on conditional density estimation has been accepted to IEICE Transactions.
  • Our papers on density ratio estimation, speaker identification, and audio tagging have been accepted to ICASSP2010.
  • Our paper on high-dimensional density-ratio estimation has been accepted to SDM2010.
  • Our paper on brain-computer interfaces has been accepted to IEEE Transactions on Biomedical Engineering.
  • Our paper on outlier detection has been accepted to Knowledge and Information Systems.
  • Our paper on active learning in reinforcement learning has been accepted to Neural Networks.
  • 1st: We give a talk on dimensionality reduction at Database Section of Strategic Research Program for Brain Sciences.
  • 3rd: We give a talk on density ratio estimation at CompView GCOE Symposium.
  • 4th: We present our work on age prediction from face images at ViEW2009.
  • 12th: We present our work on large-scale sparse learning at OPT2009.
  • 17th: We present our works on observational reinforcement learning and conditional density estimation at IPSJ SIG MPS.
  • 17th: We give a talk on pattern recognition at IEICE PRMU.
  • 18th: Year-end party is held.
Nov. 2009
  • 3rd: We give a talk on density ratio estimation at ACML2009.
  • 5th: We give a talk on active learning at Nanjing University.
  • 10th: We give a talk on active learning at The University of Tokyo.
  • 26th: We give a talk on active learning at Nara Institute of Science and Technology (NAIST).
  • 27th: We give a talk on density ratio estimation at National Institute of Information and Communications Technology (NICT).
  • Our paper on large-scale sparse learning has been accepted to OPT2009.
Oct. 2009
  • Hirotaka Hachiya (postdoc), Rubaiya Khan (master course), and Amir Sattari (YSEP) joined Sugiyama Lab.
  • Mr. Paul von Bünau (Technical University of Berlin, Berlin, Germany) is staying with us for a month.
  • 16th: We give a talk on density ratio estimation at NLP Joint Seminar.
  • 17th: Joint alumni party of Iijima-Ogawa-Sugiyama Lab. is held.
  • 19-21th: At IBIS2009, we present our works on observational reinforcement learning, risk-sensitive reinforcement learning, multi-task learning, density ratio estimation with dimensionality reduction, and semi-supervised canonical correlation analysis.
  • 24-25th: Open lab at the campus festival.
Sep. 2009
Aug. 2009
Jul. 2009
Jun. 2009
May 2009
  • Our paper on active learning in collaborative filtering has been accepted to IPSJ Transaction.
  • Our paper on stability of density-ratio estimation has been accepted to NUMML2009.
  • Our paper on optical surface profiling has been accepted to Applied Optics.
  • 13th: We have a 5th-floor joint party.
  • 26-27th: We have a seminar trip in Ohyama.
Apr. 2009
Mar. 2009
  • Our paper on nu-support vector machines has been accepted to New Generation Computing.
  • Our paper on mutual information estimation has been accepted to ISIT2009.
  • Tatsuya Yokota (bachelor 4th) received the Best Presentation Award from Department of International Development Engineering, Tokyo Institute of Technology.
  • Our research proposal was accepted to Asian Office of Aerospace Research and Development (AOARD).
  • 19th: Mr. Yu Takahashi (Nara Institute of Science and Technology) is visiting us and giving a talk on speech processing.
  • 26th: Neil Rubens received the D.E. degree, and Masahiro Koujima and Tatsuya Yokota received the B.E. degree.
Feb. 2009
  • Our paper on large-scale dimensionality reduction has been accepted to IEICE Transactions.
  • Our paper on probabilistic principal component analysis has been accepted to IJCNN2009.
  • 27th: We have a graduation party.
Jan. 2009
Dec. 2008
  • Our papers on link prediction and change-point detection have been accepted to SDM2009.
  • Our paper on off-policy reinforcement learning has been accepted to IEEE ADPRL2009.
  • Our paper on matrix factorization has been accepted to PAKDD2009.
  • Our paper on speaker identification has been accepted to ICASSP2009.
  • Our paper on independent component analysis has been accepted to ICA2009.
  • Our paper on pool-based active learning has been accepted to Machine Learning.
  • 20th: Alumni party is held.
  • 22nd: Year-end party is held.
Nov. 2008
Oct. 2008
Sep. 2008
Aug. 2008
  • Mr. Ron Begleiter (Technion Israel Institute of Technology) is staying with us for two months, and gives a talk on active learning on 20th.
  • A book on reinforcement learning has been published from Mainichi Communications.
Jul. 2008
Jun. 2008
May 2008
  • Makoto Yamada (doctor course) joined Sugiyama Lab.
  • Masashi Sugiyama (associate professor) received the 21st Ando Hiroshi Memorial Award from the Foundation of Ando Laboratory.
  • Our paper on density-ratio based pattern recognition has been accepted to MIRU2008.
  • 9th: At Information Physics Seminar of Tohoku University, Masashi Sugiyama (associate professor) gives a seminar talk on supervised learning under non-stationary environment.
  • 19th: Dr. Muneki Yasuda (Tohoku University) is visiting us and giving a talk on approximate learning rule for Boltzmann machines.
Apr. 2008
  • Yoshinobu Kawahara (post-doc researcher), Ryota Tomioka (post-doc researcher), Yuta Itoh (master course), Satoshi Hara (master course), Masahiro Koujima (bachelor course), Tatsuya Yokota (bachelor course), and Masako Nakazawa (secretary) joined Sugiyama Lab.
  • Our paper on off-policy reinforcement learning has been accepted to AAAI2008.
  • Our paper on theoretical analysis of nu-support vector machine has been accepted to ICML2008.
  • Our paper on generalization error bound for boosting has been accepted to COLT2008.
  • 3rd: Dr. Jun-ya Gotoh (Chuo University) is visiting us and giving a talk on portfolio selection.
  • 28th: At IBM T. J. Watson Research Center (Hawthorne), Masashi Sugiyama (associate professor) gives a seminar talk on covariate shift adaptation.
  • 29th: At IBM T. J. Watson Research Center (Yorktown), Masashi Sugiyama (associate professor) gives a seminar talk on covariate shift adaptation.
Mar. 2008
Feb. 2008
Jan. 2008
Dec. 2007
Nov. 2007
Oct. 2007
Sep. 2007
Aug. 2007
July 2007
June 2007
May 2007
  • 18th: We hold the 3rd T-PRIMAL seminar.
Apr. 2007
Mar. 2007
Feb. 2007
Jan. 2007
Dec. 2006
Nov. 2006
Oct. 2006
  • Hirotaka Hachiya (doctor course) and Tsubasa Fukuyama (bachelor course) joined us.
Sep. 2006
Aug. 2006