I'm the Head of
Media Analytics Department of NEC Labs in Silicon Valley,
California, leading the development of
intelligent systems for machine learning, image recognition, multimedia search,
video surveillance, recommendation, data mining, and human-computer interface. Before joining NEC, I was a senior research scientist at Siemens.
I obtained PhD in Computer Science at
University of Munich, Germany.
2011 1.
Contextual Weighting for Vocabulary Tree based Image
Retrieval [pdf] Xiaoyu Wang, Ming Yang, Timothee Cour,
Shenghuo
Zhu, Kai Yu, and Tony X. Han To Appear in Proceedings of IEEE Conference on Image
Processing (ICCV 2011) 2. Real-time clothing
recognition in surveillance videos [pdf] Ming Yang
and Kai Yu To Appear
in Proceedings of IEEE Conference on Image Processing (ICIP 2011) 3. Correspondence
Driven Adaptation for Human Profile Recognition [pdf]
Ming Yang,
Shenghuo
Zhu, Fengjun
Lv,
and Kai Yu Proceedings
of IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2011) Improve any surveillance system without any labeling 4. Large-scale Image
Classification: Fast Feature Extraction and SVM Training [pdf]
Yuanqing Lin, Fengjun Lv,
Shenghuo
Zhu, Ming Yang, Timothee Cour,
Kai Yu, Liangliang
Cao, and Thomas Huang Proceedings of IEEE Conference on Computer Vision
and Pattern Recognition (CVPR 2011) Want to know how we won ImageNet
2010? 5. Learning Image
Representations from the Pixel Level via Hierarchical Sparse Coding [pdf]
Kai Yu, Yuanqing Lin,
and John Lafferty Proceedings of IEEE Conference on Computer Vision
and Pattern Recognition (CVPR 2011) A two-layer sparse coding model that beats
sparse coding on SIFT … 2010 6. Deep Coding
Networks [pdf] Yuanqing Lin,
Tong Zhang, Shenghuo Zhu, and Kai Yu Advances in Neural Information Processing
Systems 23 (NIPS 2010) Let LCC go deeper … 7. Predicting Facial
Beauty without Landmarks [pdf]
Douglas Gray, Kai Yu, Wei Xu,
and Yihong
Gong Proceedings of the 11th European Conference
on Computer Vision (ECCV 2010) A fully
automatically trained model to tell how attractive you are … 8. Image
Classification using Super-Vector Coding of Local Image Descriptors [pdf] Xi Zhou, Kai Yu, Tong Zhang, and Thomas
Huang Proceedings of the 11th European Conference
on Computer Vision (ECCV 2010) You may not know,
this is in fact a special case of the local tangent stuff. 9. Efficient Highly
Over-Complete Sparse Coding using a Mixture Model [pdf] Jianchao Yang,
Kai Yu, and Thomas Huang Proceedings of the 11th European Conference
on Computer Vision (ECCV 2010) Making sparse
coding a bit more complex … 10. Improved Local
Coordinate Coding using Local Tangents [pdf] Kai Yu, Tong Zhang Proceedings of the 27th International
Conference on Machine Learning (ICML 2010) Okay, a simple way to further improve LCC,
and even sparse coding. 11. 3D Convolutional
Neural Networks for Human Action Recognition [pdf] Shuiwang Ji,
Wei Xu,
Ming Yang, and Kai Yu Proceedings of the 27th International
Conference on Machine Learning (ICML 2010) First such fully
trained deep model for action recognition. 12. Supervised
Translation-Invariant Sparse Coding [pdf] Jianchao Yang,
Kai Yu, Thomas Huang Proceedings of IEEE Conference on Computer
Vision and Pattern Recognition (CVPR 2010) Supervised training
to further improve sparse coding for classification 13. Learning
Locality-constrained Linear Coding for Image Classification [pdf] Jingjun Wang, Jianchao
Yang, Kai Yu, Fengjun Lv,
Thomas Huang, and Yihong Gong Proceedings of IEEE Conference on Computer
Vision and Pattern Recognition (CVPR 2010) A fast & simplified
implementation of LCC that leads to state-of-the-art performance. 2009 14. Nonlinear Learning
using Local Coordinate Coding [pdf] [appendix] Kai Yu, Tong Zhang, and Yihong
Gong Advances in Neural Information Processing
Systems 22 (NIPS 2009) A longer version in Technical Report arXiv:0906.5190v1
[pdf] A theoretical perspective
trying to understand sparse coding and develop a better one. 15. UCSC at Relevance
Feedback Track [pdf] Lanbo Zhang, Yi
Zhang, Jadiel
de Arma,
and Kai Yu In Proceedings of the 17th Text REtrieval
Conference (TREC), Gaithersburg, USA, 2009 (TREC 2009) We are No.2
position. Active learning using Transductive
Experimental Design rocks! 16. Human Action
Detection by Boosting Efficient Motion Features [pdf] Ming Yang, Fengjun
Lv,
Wei Xu,
Kai Yu, Yihong
Gong..
IEEE Workshop on Video-oriented Object and
Event Classification in Conjunction with ICCV (VOEC'2009) See how we achieve
the best performance of action recognition in TRECVID! 17. Detecting Video
Events Based on Action Recognition in Complex Scenes using Spatio-temporal
Descriptor [pdf] Guangyu Zhu,
Ming Yang, Kai Yu, Wei Xu, Yihong
Gong. ACM International Conference on Multimedia
(ACM MM'2009) 18. Large-scale
Collaborative Prediction Using a Nonparametric Random Effects Model [pdf] [slides] Kai Yu, John Lafferty, Shenghuo
Zhu, and Yihong
Gong Proceedings of the 26th International
Conference on Machine Learning (ICML 2009) A note on inverted Wishart distributions. Beyond matrix
factorization – use user and item attributes! 19. Fast Nonparametric
Matrix Factorization for Large-scale Collaborative Filtering [pdf] [code] Kai Yu, Shenghuo
Zhu, John Lafferty, and Yihong Gong Proceedings of the 32nd Annual
International ACM SIGIR Conference (SIGIR 2009) One of the best
matrix factorization you can find, in accuracy and efficiency! 20. Linear Spatial
Pyramid Matching Using Sparse Coding for Image Classification [more info] Jianchao Yang,
Kai Yu, Yihong
Gong, and Thomas Huang IEEE Conference on Computer Vision and
Pattern Recognition (CVPR 2009) Can’t believe,
linear SVM + Sparse Coding = 73% on Caltech-101! 21. Deep Learning with
Kernel Regularization for Visual Recognition [pdf] Kai Yu, Wei Xu,
and Yihong
Gong Advances in Neural Information Processing
Systems 21 (NIPS 2008) (Eds.) D. Koller,
Y. Bengio,
D. Schuurmans
and L. Bottou,
MIT Press, Cambridge, MA, USA, 2009 22. Stochastic
Relational Models for Large-scale Dyadic Data using MCMC [pdf] [code] Shenghuo Zhu,
Kai Yu and Yihong
Gong Advances in Neural Information Processing
Systems 21 (NIPS 2008) (Eds.) D. Koller,
Y. Bengio,
D. Schuurmans
and L. Bottou,
MIT Press, Cambridge, MA, USA, 2009 2008 23. Training
Hierarchical Feed-forward Visual Recognition Models Using Transfer Learning
from Pseudo Tasks [pdf] Amr Ahmed, Kai
Yu, Wei Xu,
Yihong
Gong and Eric P. Xing The 10th European Conference on Computer
Vision, (ECCV 08) Oral presentation, acceptance rate
4.6% 24. Feature Selection
for Gene Expression using Model-based Entropy [draft] Shenghuo
Zhu, Dingding Wang, Kai Yu, Tao
Li, and Yihong Gong. IEEE/ACM Transactions on Computational
Biology and Bioinformatics, 2008 25. Nonparametric
Relational Learning for Social Network Analysis [pdf] Zhao Xu, Volker Tresp, Shipeng Yu, and Kai Yu. The 2nd KDD workshop on Social Network
Mining and Analysis, (SNA-KDD 08) 26. Non-greedy Active
Learning for Text Categorization using Convex Transductive
Experimental Design [pdf] Kai Yu, Shenghuo
Zhu, Wei Xu,
and Yihong
Gong Proceedings of the 31st Annual
International ACM SIGIR Conference (SIGIR 08) 27. Learning Multiple
Graphs for Document Recommendations [pdf] Ding Zhou, Shenghuo
Zhu, Kai Yu, Xiaodan Song, Belle Tseng, Hongyuan
Zha,
and C.Lee
Giles Proceedings of the 17th International World
Wide Web Conference (WWW08) 28. Gaussian Process
Models for Link Analysis and Transfer Learning [pdf] Kai Yu and Wei Chu Advances in Neural Information Processing
Systems 20 (NIPS 2007) (Eds.) Platt, J. C., D. Koller,
Y. Singer, S. Roweis, MIT Press, Cambridge, MA,
USA, 2008 29. Predictive Matrix-Variate
t Models [pdf] Shenghuo Zhu,
Kai Yu and Yihong
Gong Advances in Neural Information Processing
Systems 20 (NIPS 2007) (Eds.) Platt, J. C., D. Koller,
Y. Singer, S. Roweis, MIT Press, Cambridge, MA,
USA, 2008 2007 30. Robust Multi-Task
Learning with t-Processes [pdf]
Shipeng Yu,
Volker Tresp,
Kai Yu, and Bharat Rao Proceedings of the 24th International
Conference on Machine Learning (ICML 2007) 31. Local Learning
Projection [pdf] Mingrui Wu, Kai
Yu, Shipeng
Yu, and Bernhard Schölkopf Proceedings of the 24th International
Conference on Machine Learning (ICML 2007) 32. Combining Contents
and Links for Classification using Matrix Factorization [pdf] Shenghuo Zhu,
Kai Yu, Yun
Chi, and Yihong
Gong Proceedings of the 30th Annual
International ACM SIGIR Conference (SIGIR 2007) 33. Stochastic
Relational Models for Discriminative Link Prediction [pdf] [data] Kai Yu, Wei Chu, Shipeng
Yu, Volker Tresp,
and Zhao Xu Advances in Neural Information Processing
Systems 19 (NIPS 2006), (Eds.) B. Schölkopf,
J. C. Platt and T. Hofmann, MIT Press, Cambridge, MA USA, 2007 2006 34. Multi-Output
Regularized Feature Projection [Link] Shipeng
Yu, Kai Yu, Volker Tresp, and Hans-Peter Kriegel.
IEEE Transactions on Knowledge and Data
Engineering (TKDE), 18 (22), pp. 1600-1613, December 2006. 35. Variational Bayesian Dirichlet-Multinomial
Allocation for Exponential Family Mixtures
Shipeng Yu, Kai Yu, and Volker Tresp, Proceedings of the 17th European Conference
on Machine Learning (ECML 2006),
2006 36. Supervised
Probabilistic Principal Component Analysis
[pdf] Shipeng Yu, Kai
Yu, Volker Tresp,
Hans-Peter Kriegel,
and Mingrui
Wu Proceedings of the 12th International
Conference on Knowledge Discovery and Data Mining (SIGKDD 2006),
2006 (Acceptance Rate <11%) 37. Infinite Hidden
Relational Models [pdf] Zhao Xu, Volker Tresp,
Kai Yu, and Hans-Peter Kriegel, Proceedings of the 22nd International
Conference on Uncertainty in Artificial Intelligence (UAI 2006),
2006 38. Active Learning via
Transductive
Experimental Design [pdf] [matlab
code & data] Kai Yu, Jinbo
Bi, and Volker Tresp, Proceedings of the 23rd International Conference
on Machine Learning (ICML 2006), 2006
39. Collaborative
Ordinal Regression [pdf] Shipeng Yu, Kai
Yu, and Volker Tresp, Proceedings of the 23rd International Conference
on Machine Learning (ICML 2006), 2006 40. Soft Clustering on
Graphs [pdf] Kai Yu, Shipeng
Yu, and Volker Tresp, Advances in Neural Information Processing
Systems 18 (NIPS 2005), (Eds.) Y. Weiss, B. Schölkopf,
and J. Platt, MIT Press, Cambridge, MA USA, 2006 2005 41. Learning to Learn
and Collaborative Filtering [pdf] Kai Yu and Volker Tresp, Workshop on Inductive Transfer: 10 Years
Later (NIPS*05 Workshop), Whistler, Canada, Dec. 2005 42. A Probabilistic
Clustering-Projection Model for Discrete Data [pdf] Shipeng Yu, Kai
Yu, Volker Tresp,
and Hans-Peter Kriegel, Proceedings of the 9th European Conference
on Principles and Practice of Knowledge Discovery in Databases (PKDD 2005),
Porto, Portugal, October 3-7, 2005. (Best
Paper Runner-up Award) 43. Learning Gaussian
Processes from Multiple Tasks [pdf] [slides] Kai Yu, Volker Tresp,
and Anton Schwaighofer, Proceedings of the 22nd International
Conference on Machine Learning (ICML 2005), Bonn, Germany, 7-11 August, 2005 44. Dirichlet Enhanced
Probabilistic Relational Model [pdf] Zhao Xu, Volker Tresp,
Kai Yu, Shipeng
Yu, and Hans-Peter Kriegel, Proceedings of the 22nd International
Conference on Machine Learning (ICML 2005), Bonn, Germany, 7-11 August, 2005 45. Blockwise Supervised
Inference on Large Graphs [pdf] Kai Yu, Shipeng
Yu, and Volker Tresp, Workshop on Learning with Partially
Classified Training Data (ICML 2005 Workshop), at the 22nd
International Conference on Machine Learning, Bonn, Germany, 7-11 August, 2005 46. Multi-Output
Regularized Projection [pdf] Kai Yu, Shipeng
Yu, and Volker Tresp, Proceedings of International IEEE
Conference on Computer Vision and Pattern Recognition (CVPR 2005), San Diego, CA, USA, June 20-26, 2005. 47. Learning Gaussian
Process Kernels via Hierarchical Bayes [pdf] Anton Schwaighofer,
Volker Tresp,
and Kai Yu, Advances in Neural Information Processing
Systems 17 (NIPS 2004). (Eds.) Saul, L.K., Y. Weiss and L. Bottou, MIT Press,
Cambridge, MA, USA, 2005. 48. Dirichlet Enhanced Latent
Semantic Analysis [pdf] Kai Yu, Shipeng
Yu, and Volker Tresp, Proceedings of Artificial Intelligence
& Statistics (AISTATS 2005), Barbados, January 6-8, 2005. 49. Multi-Label
Informed Latent Semantic Indexing [pdf] Kai Yu, Shipeng
Yu, and Volker Tresp, Proceedings of the 28th International ACM SIGIR
Conference on Research and Development in Information Retrieval (SIGIR 2005),
August 15-19, 2005, in Salvador,
Brazil. 2004 50. A Nonparametric
Hierarchical Bayesian Framework for Information Filtering [pdf] Kai Yu, Volker Tresp,
and Shipeng
Yu, Proceedings of the 27th International ACM
SIGIR Conference on Research and Development in Information Retrieval (SIGIR
2004), Sheffield, UK, July 25 - 29, 2004. 51. An introduction to
nonparametric hierarchical Bayesian modelling with
a focus on multi-agent learning [pdf] Volker Tresp and Kai Yu.
book chapter,
Switching and Learning in Feedback Systems. Springer, 2004.
52. Probabilistic
Memory-Based Collaborative Filtering [draft] Kai Yu, Anton Schwaighofer,
Volker Tresp,
Xiaowei
Xu,
Hans-Peter Kriegel
IEEE Transactions on Knowledge and
Data Engineering (TKDE), Vol.16, No.1, pp. 56--69, 2004. 2003 53. Approximate
Solutions to Nonparametric Bayesian Hierarchical Modelling
with Applications to Information Filtering [link] Volker Tresp,
Kai Yu, Anton Schwaighofer, Nonparametric Bayesian Methods and Infinite
Models, (NIPS*03 Workshop), Vancouver, Dec. 2003. 54. Collaborative
Ensemble Learning: Combining Collaborative and Content-Based Information
Filtering via Hierarchical Bayes [pdf] Kai Yu, Anton Schwaighofer,
Volker Tresp,
Wei-Ying Ma, Hongjiang Zhang, Proceedings of the 19th International
Conference on Uncertainty in Artificial Intelligence (UAI 2003), Acapulco, Mexico, August 7-10, 2003.
(Plenary oral presentation, 11% accepted) 55. Knowing a Tree from
the Forest: Art Image Retrieval using a Society of Profiles [pdf] Kai Yu, Wei-Ying Ma, Volker Tresp,
Zhao Xu,
Xiaofei
He, Hongjiang
Zhang and Hans-Peter Kriegel, Proceedings of the 11th Annual ACM
International Conference on Multimedia (ACM Multimedia 2003), Berkeley, CA, USA, November 2-8, 2003.
Selected
Publications