Google Scholar Research Gate Github


    2017 ~ Present

  • Ming-Fang Chang, John Lambert, Patsorn Sangkloy, Jagjeet Singh, Slawomir Bak, Andrew Hartnett, De Wang, Peter Carr, Simon Lucey, Deva Ramanan, James Hays. Argoverse: 3D Tracking and Forecasting With Rich Maps. CVPR 2019. [pdf] [site]
  • Jack W. Stokes, De Wang, Mady Marinescu, Marc Marino, Brian Bussone. Attack and Defense of Dynamic Analysis-Based, Adversarial Neural Malware Classification Models. ArXiv preprint. [pdf]
  • Bin Gu, De Wang, Zhouyuan Huo, Heng Huang. Inexact Proximal Gradient Methods for Non-convex and Non-smooth Optimization. AAAI 2018.
         Proposed a proximal gradient algorithm for non-convex, non-smooth optmization, with faster convergence rate.
  • 2016

  • De Wang, Feiping Nie, Heng Huang. Learning Task Relational Structure for Multi-Task Feature Learning. ICDM 2016 (International Conference on Data Mining). [pdf] [bibtex] [slides]
         Proposed a new regualrizing norm for uncovering group structure of tasks in multi-task learning, which helps knowledge transfer between tasks.
  • De Wang, Feiping Nie, Heng Huang. Fast Robust Non-negative Matrix Factorization for Large Scale Data Clustering. IJCAI 2016 (International Joint Conference on Artificial Intelligence). [pdf] [bibtex] [slides] [poster] [code]
  • 2015

  • De Wang, Feiping Nie, Heng Huang. Feature Selection via Global Redundancy Minimization. IEEE T-KDE (Transaction on Knowledge and Data Engineering). [pdf] [bibtex] [dataset] [code]
         Proposed a general framework for re-ranking features obtained by any filter feature selection methods (both supervised and unsupervised methods) in order to achieve minimum redundancy in top ranked features.
  • 2014

  • De Wang, Feiping Nie, Heng Huang. Large Scale Adaptive Semi-supervised Learning via Unified Inductive and Transductive Model. SIGKDD 2014, Oral Presentation (ACM SIGKDD Conference on Knowledge Discovery and Data Mining). [pdf] [bibtex] [KDD MadnessTrailer] [Slides] [Code]
         Avoids computational expensive (at least O(n^2)) step of constructing Laplacian graph in SSL, reduce the computational complexity to linear to sample number n, scalable to large data.
  • De Wang, Feiping Nie, Heng Huang. Unsupervised Feature Selection via Unified Trace Ratio Formulation and K-means Clustering (TRACK). ECML/PKDD 2014 (European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases). [pdf] [bibtex] [Code]
         Proposed a unified framework for extending trace ratio criterion to unsupervised circumstance, which seamlessly combines k-means clustering and trace ratio dimensionality reduction. It can be used for unsupervised dimensionality reduction, clustering, and feature selection.
  • De Wang, Yang Wang, Feiping Nie, Jingwen Yan, Andew Saykin, Li Shen, Heng Huang. Human Connectome Module Pattern Detection Using A New Multi-Graph MinMax Cut Model. MICCAI 2014 (International Conference on Medical Image Computing and Computer Assisted Intervention). [pdf] [bibtex] [Poster]
         Extends min-max cut to multi-view/multi-modality/multi-graph circumstance to discover consistent strongly connected module in human brain.
  • 2013

  • De Wang, Feiping Nie, Heng Huang, Jingwen Yan, Shannon Risacher, Andrew Saykin, Li Shen. Structural Brain Network Constrained Neuroimaging Marker Identification for Predicting Cognitive Functions. IPMI 2013 (International Conference on Information Processing in Medical Imaging). [pdf] [bibtex] [Poster] [Code]
         Proposed a framework for incorporating prior feature/bio-marker correlation graph information for selecting correlated features which has synergic effect towards a certain human function.
  • Yang, Liang, Zheng-Qi Fu, De Wang, He-Long Li, and Jing-Bo Xia. An improved ant colony algorithm for continuous space optimization. In Machine Learning and Cybernetics (ICMLC), 2010 International Conference on, vol. 4, pp. 1829-1832. [pdf] [bibtex]
         Improved ant colony algorithm by a better moving strategy and pheromone distribution.
  • De Wang, Liang Yang, Zhengqi Fu and Jingbo Xia. Prediction of Thermophilic Protein with Pseudo Amino Acid Composition: An Approach from Combined Feature Selection and Reduction. Protein & Peptide Letters, 2011, Vol. 18, No. 7, 684-689. [pdf] [bibtex]