Chun-Ta Lu / 盧俊達

I'm a software engineer at Google AI (Research and Machine Intelligence). I received the Ph.D. in the Department of Computer Science, University of Illinois at Chicago. During the doctoral study, I worked as a research assistant in the BDSC Lab advised by Prof. Philip S. Yu. Before joining UIC, I received the B.S. degree from National Taiwan University and the M.S. degree from National Chiao Tung University.

My research interests lie in the fields of machine learning and data mining. In particular, I focus on the development and analysis of algorithms for heterogenous information networks, as well as heterogeneous data fusion for multi-task multi-view learning, deep feature representation, and relation learning.

Resume

Publications

Spectral Collaborative Filtering

Lei Zheng, Chun-Ta Lu, Fei Jiang, Jiawei Zhang and Philip S. Yu
ACM Conference On Recommender Systems (RecSys), 2018

[paper]

Learning from Multi-View Multi-Way Data via Structural Factorization Machines

Chun-Ta Lu, Lifang He, Hao Ding and Philip S. Yu
The Web Conference (WWW), 2018

[paper]

Kernelized Support Tensor Machines

Lifang He, Chun-Ta Lu, Guixiang Ma, Shen Wang, Linlin Shen, Philip S. Yu, and Ann B. Ragin
International Conference on Machine Learning (ICML), 2017

[paper]

Structural Deep Brain Network Mining

Shen Wang, Lifang He, Bokai Cao, Chun-Ta Lu, Philip S. Yu and Ann B. Ragin
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2017

(Oral presentation, acceptance 8.56%)
[paper]

Multi-way Multi-level Kernel Modeling for Neuroimaging Classification

Lifang He, Chun-Ta Lu, Hao Ding, Shen Wang, Linlin Shen, Philip S. Yu and Ann B. Ragin

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017
[paper]

Multilinear Factorization Machines for Multi-Task Multi-View Learning

Chun-Ta Lu, Lifang He, Weixiang Shao, Bokai Cao and Philip S. Yu
ACM International Conference on Web Search and Data Mining (WSDM), 2017

[paper] [slides] [poster] [code]

Multi-view Graph Embedding with Hub Detection for Brain Network Analysis

Guixiang Ma, Chun-Ta Lu, Lifang He, Philip S. Yu and Ann Ragin
IEEE International Conference on Data Mining (ICDM), 2017

[paper]

A Broad Learning Approach for Context-Aware Mobile Application Recommendation

Tingting Liang, Lifang He, Chun-Ta Lu, Liang Chen, Philip S. Yu and Jian Wu
IEEE International Conference on Data Mining (ICDM), 2017

[paper]

Multi-view Clustering via Graph Embedding for Connectome Analysis

Guixiang Ma, Lifang He, Chun-Ta Lu, Weixiang Shao, Philip S. Yu, Ann Ragin and Alex Leow
International Conference on Information and Knowledge Management (CIKM), 2017

[paper]

Collective Geographical Embedding for Geolocating Social Network Users

Fengjiao Wang, Chun-Ta Lu, Yongzhi Qu, and Philip S. Yu
Pacific Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2017

[paper]

HEER: Heterogeneous Graph Embedding for Emerging Relation Detection from News

Jingyuan Zhang, Chun-Ta Lu, Mianwei Zhou, Sihong Xie, Yi Chang, and Philip S. Yu
IEEE International Conference on Big Data, 2016

[paper] [slides]

Online Multi-view Clustering with Incomplete Views

Weixiang Shao, Lifang He, Chun-Ta Lu and Philip S. Yu
IEEE International Conference on Big Data, 2016

[paper] [slides] [code]

Interpretable and Effective Opinion Spam Detection via Temporal Patterns Mining across Websites

Yuan Yuan, Sihong Xie, Chun-Ta Lu, Philip S. Yu, and Jie Tang
IEEE International Conference on Big Data, 2016

[paper] [slides]

Community Detection with Partially Observable Links and Node Attributes

Xiaokai Wei, Bokai Cao, Weixiang Shao, Chun-Ta Lu and Philip S. Yu
IEEE International Conference on Big Data, 2016

[paper] [slides]

Online Unsupervised Multi-view Feature Selection

Weixiang Shao, Lifang He, Chun-Ta Lu, Xiaokai Wei and Philip S. Yu
IEEE International Conference on Data Mining (ICDM), 2016

[paper] [slides] [code] [bib]

Semi-supervised Tensor Factorization for Brain Network Analysis

Bokai Cao, Chun-Ta Lu, Xiaokai Wei, Philip S. Yu, and Alex D. Leow
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML/PKDD), 2016

[paper] [slides] [code] [bib]

Joint Community and Structural Hole Spanner Detection via Harmonic Modularity

Lifang He, Chun-Ta Lu, Jiaqi Ma, Jianping Cao, Linlin Shen, and Philip S. Yu
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2016
(Oral presentation, acceptance 8.93%)

[paper] [video] [slides] [code] [bib]

Item Recommendation for Emerging Online Businesses

Chun-Ta Lu, Sihong Xie, Weixiang Shao, Lifang He, and Philip S. Yu
International Joint Conference on Artificial Intelligence (IJCAI), 2016

[paper] [slides] [poster] [bib]

Spatio-Temporal Tensor Analysis for Whole-Brain fMRI Classification

Guixiang Ma, Lifang He, Chun-Ta Lu, Philip S. Yu, Linlin Shen, and Ann B. Ragin
SIAM International Conference on Data Mining (SDM), 2016

[paper] [slides] [poster] [bib]

Identifying Connectivity Patterns for Brain Diseases via Multi-side-view Guided Deep Architectures

Jingyuan Zhang, Bokai Cao, Sihong Xie, Chun-Ta Lu, Philip S. Yu, and Ann B. Ragin
SIAM International Conference on Data Mining (SDM), 2016

[paper] [slides] [poster] [bib]

Identifying your Customers in Social Networks

Chun-Ta Lu, Hong-Han Shuai, and Philip S. Yu
ACM International Conference on Information and Knowledge Management (CIKM), 2014

[paper] [slides] [data] [bib]

Inferring the Impacts of Social Media on Crowdfunding

Chun-Ta Lu, Sihong Xie, Xiangnan Kong, and Philip S. Yu
ACM International Conference on Web Search and Data Mining (WSDM), 2014

[paper] [slides] [poster] [data] [bib]

Exploring Application Usage Patterns of Smart Phones for Discovering Personal Semantic Regions

Chun-Ta Lu
Master’s Thesis, 2012

[paper] [slides]

A Framework of Mining Semantic Regions from Trajectories

Chun-Ta Lu, Po-Ruey Lei, Wen-Chih Peng, and Ing-Jiunn Su
International Conference on Database Systems for Advanced Application (DASFAA), 2011

[paper] [slides] [bib]

Professional Experience

Software Engineering Intern, Google Research
May. 2017 - Aug. 2017

I developed deep semi-supervised neural networks for multi-label image classification. It invovles constructing an End-To-End training/serving pipeline for large-scale binary classifiers built upon Tensorflow and Flume (an internal data-parallel pipeline). It is able to train 17M images annotated with labels spanning over 20K categories within a few hours. I also designed and implemented Tensorflow custom Ops for enabling a smooth distribution of embeddings for related instances and/or labels in neural networks.

Research Assistant, UIC
May 2013 - Present

I analyze the hidden connections between the fundraising results of projects on crowdfunding websites and the corresponding promotion campaigns in social media. I further developed a methodology CSI (Customer-Social Identification) for identifying customers in online social networks effectively by using the basic information of customers, such as username and purchase history. After connecting customers (and also products) in multiple websites (e.g., Amazon and Twitter), I proposed a personalized recommendation algorithm that utilizes complimentary information from multiple websites to improve recommendation performance.
Currently, I am developing tensor-based factorization machines for multi-task multi-view learning. I also working on deep multi-view factorization machines that can simultaneously learn deep feature representations and relational structures from multi-view data.

Quantitative Analyst Intern, Google
Jun. 2014 - Aug. 2014

I analyzed the correlations between the signals (e.g., CTR, CPC, RPM and quality scores) and the changes of signals of AdWords. I analyzed the effectiveness of AdWords campaign treatments through the changes of signals.

Web Developer, BIKE ID Co., Ltd
Jan. 2012 – Aug. 2012

I developed a website for Taipei city public bicycle services, which can generate customized bicycle tour brochures. I developed a website and a Facebook App for Taiwan running community.

Research Assistant, NCTU
Sep. 2010 – Jan. 2012

I proposed a framework for extracting semantic regions from user behaviors, which has been published in DASFAA'10. I explored Android App usage patterns relative to location dependency and analyzed semantics of regions from geographic information and App usage patterns. The results are published in my Master's Thesis. This RA position is sponsored by NCTU and also High Tech Computer Corporation (HTC)

Webmaster
Mar. 2010 – Nov. 2010

I designed the website for ACM SIGSPATIAL International Workshop on Location-Based Social Networks (LBSN), 2010. The templates are also used in the following years. I also designed the website for Conference on Technologies and Applications of AI (TAAI), 2010.

Research Assistant, Academia Sinica
Jul. 2008 – Aug. 2009

I help developed mobile medication administration tool for hospital service. In this project, I designed the report generator for patient records and medical supply inventory records. I also help developed the user interface for an intelligent mobile nursing cart (iNuC).