Research

Stochastic Optimization for Large-Scale Machine Learning

  • "VR-SGD: A Simple Stochastic Variance Reduction Method for Machine Learning" (with Kaiwen Zhou, Hongying Liu, James Cheng, Ivor W. Tsang, Lijun Zhang, Dacheng Tao, Licheng Jiao). To appear in IEEE Transactions on Knowledge and Data Engineering, 2018. [Preprint] Code
  • "Accelerated Proximal Stochastic Variance Reduced Gradient Optimization with Momentum and Iterate Averaging" (with Yuanyuan Liu, James Cheng, Jiacheng Zhuo and Junhui Cai), Submitted, 2017. [Preprint]
  • "Fast Stochastic Variance Reduced Gradient Method with Momentum Acceleration for Machine Learning" (with James Cheng and Jiacheng Zhuo), Submitted, 2017. [Preprint]
  • "ASVRG: Accelerated Proximal SVRG" (with Licheng Jiao, Kaiwen Zhou, James Cheng, Yan Ren, Yufei Jin), to appear in Proceedings of Machine Learning Research, 2018. [PDF]
  • "A Simple Stochastic Variance Reduced Algorithm with Fast Convergence Rates" (with Kaiwen Zhou, James Cheng), in Proceedings of the 35th International Conference on Machine Learning (ICML), Stockholm, Sweden, pp. 5975–5984, 2018.
  • "Guaranteed Sufficient Decrease for Stochastic Variance Reduced Gradient Optimization" (with Yuanyuan Liu, Kaiwen Zhou, James Cheng, Kelvin Kai Wing Ng, Yuichi Yoshida), in Proceedings of the 21st International Conference on Artificial Intelligence and Statistics (AISTATS), pp. 1027–1036, 2018. [PDF]
  • "Accelerated First-order Methods for Geodesically Convex Optimization on Riemannian Manifolds" (with Yuanyuan Liu, ames Cheng, Hong Cheng, Licheng Jiao), in Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS), pp. 4875–4884, 2017.
  • "Accelerated Variance Reduced Stochastic ADMM" (with Yuanyuan Liu and James Cheng), in Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI), pp. 2287–2293, 2017.
  • "LFTF: A Framework for Efficient Tensor Analytics at Scale" (with Fan Yang, Yuzhen Huang, James Cheng, Jinfeng Li, Yunjian Zhao and Ruihao Zhao), in: Proceedings of the 43rd International Conference on Very Large Data Bases (VLDB), pp. 745–756, 2017.

Schatten Quasi-Norm Minimization

  • "Bilinear Factor Matrix Norm Minimization for Robust PCA: Algorithms and Applications" (with James Cheng, Yuanyuan Liu, Zhi-Quan Luo, and Zhouchen Lin), IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(9): 2066–2080, 2018.
  • "Fuzzy Double Trace Norm Minimization for Recommendation Systems" (with Yuanyuan Liu, James Cheng, and Da Yan), IEEE Transactions on Fuzzy Systems, 26(4): 2039-2049, 2018.
  • "LRR for Subspace Segmentation via Tractable Schatten-p Norm Minimization and Factorization" (with Hengmin Zhang, Jian Yang, Chen Gong, and Zhenyu Zhang), to appear in IEEE Transactions on Cybernetics, 2018.
  • "Unified Scalable Equivalent Formulations for Schatten Quasi-Norms" (with Yuanyuan Liu, and James Cheng), CUHK Technical Report CSE-ShangLC20160307, March 7, 2016. [Preprint]
  • "Tractable and Scalable Schatten Quasi-Norm Approximations for Rank Minimization" (with Yuanyuan Liu and James Cheng), in: Proceedings of the 19th International Conference on Artificial Intelligence and Statistics (AISTATS), pp. 620–629, 2016.
  • "Scalable Algorithms for Tractable Schatten Quasi-Norm Minimization" (with Yuanyuan Liu and James Cheng), in: Proceedings of the 30th AAAI Conference on Artificial Intelligence (AAAI), pp. 2016–2022, 2016.

Tensor Decomposition and Completion

  • "LFTF: A Framework for Efficient Tensor Analytics at Scale" (with Fan Yang, Yuzhen Huang, James Cheng, Jinfeng Li, Yunjian Zhao and Ruihao Zhao), in: Proceedings of the 43rd International Conference on Very Large Data Bases (VLDB), pp. 745–756, 2017.
  • "Generalized Higher-Order Orthogonal Iteration for Tensor Learning and Decomposition" (with Yuanyuan Liu, Wei Fan, James Cheng and Hong Cheng), IEEE Transactions on Neural Networks and Learning Systems, vol. 27, no. 12, pp. 2551–2563, 2016.
  • "Trace Norm Regularized CANDECOMP/PARAFAC Decomposition with Missing Data" (with Yuanyuan Liu, Licheng Jiao, James Cheng and Hong Cheng), IEEE Transactions on Cybernetics, vol. 45, no. 11, pp. 2437–2448, 2015.
  • "Generalized Higher-Order Orthogonal Iteration for Tensor Decomposition and Completion" (with Yuanyuan Liu, Wei Fan, James Cheng and Hong Cheng), in Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS) 27, pp. 1763–1771, 2014.
  • "Generalized Higher-Order Tensor Decomposition via Parallel ADMM" (with Yuanyuan Liu and James Cheng), in Proceedings of the 28th AAAI Conference on Artificial Intelligence (AAAI), pp. 1279–1285, 2014.
  • "Factor Matrix Trace Norm Minimization for Low-Rank Tensor Completion" (with Yuanyuan Liu, Hong Cheng, James Cheng and Hanghang Tong), in Proceedings of the 14th SIAM International Conference on Data Mining (SDM), pp. 866–874, 2014.

Matrix Completion and Recovery

  • "Robust Bilinear Factorization with Missing and Grossly Corrupted Observations" (with Yuanyuan Liu, Hanghang Tong, James Cheng, and Hong Cheng), Information Sciences, vol. 370, pp. 53–72, 2015.
  • "Nuclear Norm Regularized Least Squares Optimization on Grassmannian Manifolds" (with Yuanyuan Liu, Hong Cheng, and James Cheng), in: Proceedings of the 30th Conference on Uncertainty in Artificial Intelligence (UAI), pp. 515–524, 2014.
  • "Recovering Low-Rank and Sparse Matrices via Robust Bilateral Factorization" (with Yuanyuan Liu, James Cheng, and Hong Cheng), in: Proceedings of the 14th IEEE International Conference on Data Mining (ICDM), pp. 965–970, 2014.
  • Robust Principal Component Analysis with Missing Data (with Yuanyuan Liu, James Cheng, and Hong Cheng), in: Proceedings of the 23rd ACM International Conference on Information and Knowledge Management (CIKM), pp. 1149–1158, 2014.
  • "A Fast Tri-Factorization Method for Low-Rank Matrix Recovery and Completion" (with Yuanyuan Liu and L. C. Jiao), Pattern Recognition, 46(1): 163–173, 2013. [PDF]
  • "An Efficient Matrix Factorization Based Low-Rank Representation for Subspace Clustering" (with Yuanyuan Liu and L. C. Jiao), Pattern Recognition, 46(1): 284–292, 2013. [PDF]
  • "An Efficient Matrix Bi-Factorization Alternative Optimization Method for Low-Rank Matrix Recovery and Completion" (with Yuanyuan Liu and L. C. Jiao), Neural Networks, 48(12): 8–18, 2013. [PDF]

Semi-Supervised Learning

  • "Semi-Supervised Learning with Nuclear Norm Regularization" (with L. C. Jiao, Yuanyuan Liu, and Hanghanag Tong), Pattern Recognition, 46(8): 2323–2336, 2013. [PDF]
  • "Semi-Supervised Learning with Mixed Knowledge Information" (with L. C. Jiao and Fei Wang), in Proceedings of the 18th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), pp. 732–740, 2012. [PDF]
  • "Learning Spectral Embedding via Iterative Eigenvalue Thresholding" (with L. C. Jiao, Yuanyuan Liu, and Fei Wang), in Proceedings of the 21st ACM International Conference on Information and Knowledge Management (CIKM), pp. 1507–1511, 2012. [PDF]
  • "Learning Spectral Embedding for Semi-supervised Clustering" (with Yuanyuan Liu and Fei Wang), in Proceedings of 11th IEEE International Conference on Data Mining (ICDM), pp. 597–606, 2011. [PDF]

Dimensionality Reduction and Clustering

  • "Robust Bilinear Factorization with Missing and Grossly Corrupted Observations" (with Yuanyuan Liu, Hanghang Tong, James Cheng, and Hong Cheng), Information Sciences, vol. 370, pp. 53–72, 2015.
  • "Recovering Low-Rank and Sparse Matrices via Robust Bilateral Factorization" (with Yuanyuan Liu, James Cheng, and Hong Cheng), in: Proceedings of the 14th IEEE International Conference on Data Mining (ICDM), pp. 965–970, 2014.
  • "Fast Affinity Propagation Clustering: A Multilevel Approach" (with L. C. Jiao, Jiarong Shi, Fei Wang, and Maoguo Gong), Pattern Recognition, 45(1): 474-486, 2012.
  • "Graph Dal Regularization Non-Negative Matrix Factorization for Co-Clustering" (with L. C. Jiao and Fei Wang), Pattern Recognition, 45(6): 2237-2250, 2012.
  • "Fast Density-Weighted Low-Rank Approximation Spectral Clustering" (with L. C. Jiao, Jiarong Shi, Maoguo Gong, and R.H. Shang), Data Mining Knowledge Discovery, 23(2): 345-378, 2011. [PDF]