NAN DU

Georgia Tech, College of Computing

About Me

July 2016 - present

Aug 2011 - May 2016

  • Research Assistant, School of Computational Science & Engineering, College of Computing

    • Machine learning for user-engagement modeling, online information diffusion and maximization, and advertisement assignment over large-scale social networks.
    • I am supported by the Facebook Graduate Fellowship 2014-2015.

Contact

dunan AT google Dot com

Research

My research focuses on developing large-scale machine learning models and algorithms for statistical analysis of temporal/spatial dynamics arising from social networks and social media, which has practical applications in promoting online user engagement, optimizing advertisement allocations and providing context-aware recommendations.

Selected Publications

Conference

  • Recurrent Marked Temporal Point Processes: Embedding Event History to Vector. Nan Du, Hanjun Dai, Rakshit Trivedi, Utkarsh Upadhyay, Manuel Gomez-Rodriguez and Le Song. Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining (KDD), 2016. [PAPER]

  • Isotonic Hawkes Processes. Yichen Wang, Bo Xie, Nan Du, and Le Song. International Conference on Machine Learning (ICML), 2016, Beijing, China. [PAPER]

  • Time-Sensitive Recommendation From Recurrent User Activities . Nan Du, Yichen Wang, Niao He, and Le Song. Neural Information Processing Systems (NIPS), 2015, Montreal, Quebec, Canada. [PAPER]

  • Dirichlet-Hawke Processes with Applications to Clustering Continuous-Time Document Streams . Nan Du, Mehrdad Farajtabar, Amr Ahmed, Alexander J. Smola, and Le Song. Proceedings of the 21st ACM SIGKDD international conference on Knowledge discovery and data mining (KDD), Aug.10 - 13, 2015, Sydney, Australia. [PAPER]

  • Back to the Past: Source Identification in Diffusion Networks . Mehrdad Farajtabar, Nan Du, Mohammad Zamani, Manuel Gomez Rodriguez, and Le Song. International Conference on Artificial Intelligence and Statistics (AISTATS), 2015, California. [PAPER]

  • Learning Time-Varying Converage Functions . Nan Du, Yingyu Liang, Maria-Florina Balcan, and Le Song. Neural Information Processing Systems (NIPS), 2014, Montreal, Quebec, Canada. [PAPER][Bibtex]

  • Shaping Social Activity by Incentivizing Users. Mehrdad Farajtabar, Nan Du, Manuel Gomez Rodriguez, Isabel Valera, Hongyuan Zha, Le Song. Neural Information Processing Systems (NIPS), 2014, Montreal, Quebec, Canada. [PAPER][CODE][Bibtex]

  • Influence Function Learning in Information Diffusion Networks (full oral presentation). Nan Du, Yingyu Liang, Maria-Florina Balcan, and Le Song. International Conference on Machine Learning (ICML) , June. 22 - June 24, 2014, Beijing, China. [PAPER][CODE][Bibtex]

  • Scalable Influence Estimation in Continuous-Time Diffusion Networks (Best Paper Award, full oral presentation). Nan Du, Le Song, Manuel Gomez Rodriguez, and Hongyuan Zha. Neural Information Processing Systems (NIPS). Dec. 5 - Dec. 10, 2013, Lake Tahoe, Nevada, USA. [PAPER] [SLIDE][POSTER][CODE][Bibtex]

  • Uncover Topic-Sensitive Information Diffusion Networks (full oral presentation). Nan Du, Le Song, Hyenkyun Woo, and Hongyuan Zha. Sixteenth International Conference on Artificial Intelligence and Statistics (AISTATS) , Apr. 29 - May 1, 2013, Scottsdale, AZ, USA. [PAPER][Bibtex]

  • Learning Networks of Heterogeneous Influence (spotlight presentation). Nan Du, Le Song, Alex Smola, and Ming Yuan. Neural Information Processing Systems (NIPS) , Dec. 5 - Dec. 10, 2012, Lake Tahoe, Nevada, USA. [PAPER][Bibtex]

  • Analysis of large multi-modal social networks: patterns and a generator(full oral presentation). Nan Du, Hao Wang, and Christos Faloutsos. The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) , Sep. 20 - Sep. 24, 2010, Barcelona, Catalonia, Spain. [PAPER][Bibtex]

  • Large human communication networks: patterns and a utility-driven generator(full oral presentation). Nan Du, Christos Faloutsos, Bai Wang and Leman Akoglu. Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD) , June. 28 - July. 1, 2009, Paris, France. [PAPER][Bibtex]

  • Overlapping community structure detection in networks. Nan Du, Bai Wang and Bin Wu. Proceedings of the 17th ACM Conference on information and Knowledge Management (CIKM) , 2008, Napa Valley, California, USA. [PAPER][Bibtex]

  • Improved recommendation based on collaborative tagging behaviors. Shiwan Zhao, Nan Du, Andreas Nauerz, Xiatian Zhang, Quan Yuan, and Rongyao Fu. Proceedings of the 13th international conference on Intelligent user interfaces(IUI) , Jan 13-16, 2008, Canary Islands, Spain. [PAPER][Bibtex]

Workshop

  • Community detection in large-scale social networks. Nan Du, Bin Wu, Xin Pei, Bai Wang, and Liutong Xu. Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis (SNAKDD) , 2007, San Jose, California, USA. [PAPER][Bibtex]

  • A parallel algorithm for enumerating all maximal cliques in complex network. Nan Du, Bin Wu, Liutong Xu, Bai Wang, and Xin Pei. The Second International Workshop on Mining Complex Data, 2006, HongKong, China. [PAPER][Bibtex]

Software

PtPack is a C++ software library of high-dimensional temporal point processes. It aims to provide flexible modeling, learning, and inference of general multivariate temporal point processes to capture the latent dynamics governing the sheer volume of various temporal events arising from social networks, online media, financial trading, modern health-care, recommender systems, etc. [CODE][DOCUMENTATION]

Teaching

Teaching Assistant

Spring 2013, CSE 8803, Advanced Machine Learning, Office Hour: 2:00-3:00pm, Friday, Location: IDH 1120

Fall 2013, CSE 6740, Computational Data Analysis, Office Hour: 2:00-3:00pm, Friday, Location: IDH 1120

Skills

Programming

C/C++    Matlab  R  Java  SQL  JavaScript

HPC

MPI  OpenMP  CUDA  SSE  Hadoop

Visualization

D3  Prefuse

Machine Learning