Haofu Liao Department of Computer Science, University of Rochester, Rochester, NY (617) 775-9820 | liaohaofu@gmail.com | www.liaohaofu.com | haofu-liao-20754175 Research Interests Interested in leveraging the latest advances in deep learning and machine learning to solve problems in data mining and computer vision. Education University of Rochester Rochester, New York DOCTOR OF PHILOSOPHY IN COMPUTER SCIENCE Sep. 2015 - PRESENT Advisor: Jiebo Luo Northeastern University Boston, Massachusetts MASTER OF SCIENCE IN ELECTRICAL AND COMPUTER ENGINEERING Sep. 2012 - May 2015 • Thesis: High Performance Kernel Smoothing Library For Biomedical Imaging • Cumulative GPA: 3.9/4.0 Beijing University of Posts and Telecommunications BACHELOR OF ENGINEERING IN INFORMATION AND COMMUNICATION ENGINEERING Beijing, China Sep. 2008 - May 2012 • Thesis: Independent Component Analysis Based Blind Signal Separation • Cumulative GPA: 3.3/4.0 • Honor: Graduate with Outstanding Undergraduate Thesis Research Experience Inferring Restaurant Styles From User-Review Restaurants Rochester, New York VISTA RESEARCH GROUP, UNIVERSITY OF ROCHESTER. ADVISOR: PROF. JIEBO LUO Jan. 2016 - PRESENT • Created a restaurant photo dataset based on user-submitted photos from TripAdvisor. • Built a multi-label convolutional neural network to label user-submitted photos with restaurant style attributes. • Designed the pseudo tagging and restaurant profiling algorithms to infer restaurant styles from labeled restaurant photos. A Deep Learning Approach to Skin Disease Diagnosis VISTA RESEARCH GROUP, UNIVERSITY OF ROCHESTER. ADVISOR: PROF. JIEBO LUO Rochester, New York Sep. 2015 - PRESENT • Created a dermatology image dataset from multiple on-line skin disease atlases. To our best knowledge, our dataset, which contains 75,665 dermatology images from 2,113 skin diseases, is the largest and most comprehensive dermatology image dataset. • Built several deep convolutional neural networks to classify skin diseases, skin lesions and their anatomical locations. High-Performance Kernel Smoothing Toolbox for Biomedical Image Processing COGNITIVE SYSTEM LABORATORY, NORTHEASTERN UNIVERSITY. ADVISOR: PROF. DENIZ ERDOGMUS Boston, Massachusetts Sep. 2013 - Apr. 2015 • Studied kernel density estimation (KDE), bandwidth selection, kernel density derivative estimation (KDDE) algorithms and related applications on biomedical images. • Designed a KDE based mathematic library in C++, which included a set of univariate kernel cores, bandwidth selection methods, KDE and KDDE modules. • Accelerated the toolbox using parallel processing with both CUDA and Pthread. • Added full Matlab support for the library. Vessel Segmentation for Retinal Images and Cerebral Microvasculature Images COGNITIVE SYSTEM LABORATORY, NORTHEASTERN UNIVERSITY. ADVISOR: PROF. DENIZ ERDOGMUS Boston, Massachusetts Sep. 2013 - Apr. 2015 • Studied curvilinear structure tracing, vessel enhancement filtering, and density estimation techniques, such as principal curve tracing, Frangi filter, and kernel smoothing. • Implemented a decision-tree algorithm in Matlab, which measures the ridgeness based on the combined decision from a principal curve formula, the eigenvalues of Hessian matrix, and kernel density values. • Improved the performance of Frangi filter by introducing a kernel curvature estimator with an unconstrained bandwidth which allowed the kernel estimator to smooth in any direction whether coordinate or not. MARCH 14, 2016 HAOFU LIAO · CURRICULUM VITAE 1 Fast Level Set Method Based Image Segmentation for Luggage Inspection Boston, Massachusetts COMPUTER ARCHITECTURE RESEARCH GROUP, NORTHEASTERN UNIVERSITY. ADVISOR: PROF. DAVID KAELI • • • • Jun. 2014 - Sep. 2014 Investigated a series of level set methods and their speed function design. Improved existing fast level set method by an adaptive speed function based on Chan-Vese model. Implemented new fast level set method on GPU and optimized new speed function with parallel reduction. Optimized new algorithm by the latest Dynamic Parallelism and Hyper-Q features in CUDA. 802.11 Physical Layer Beacon Design and Implementation on USRP MATHWORKS SMART LAB, NORTHEASTERN UNIVERSITY. ADVISOR: PROF. KAUSHIK CHOWDHURY Boston, Massachusetts May 2013 - Aug. 2013 • Helped design the 802.11 PHY Layer protocol for USRP. • Designed an 802.11 WLAN transmitter and receiver in MATLAB. • Connected the 802.11 WLAN transmitter and receiver to USRP and tested their performance Work Experience University of Rochester Rochester, NY RESEARCH ASSISTANT, ADVISOR: PROF. JIEBO LUO Sep. 2015 - PRESENT • Work on the project which aims to predict attribute tags for restaurants using user-uploaded photos (See research experience in detail). • Use deep learning method to label dermatology images with skin lesion tags and help skin disease diagnosis (See research experience in detail). EMC Corporation Hopkinton, Massachusetts SYSTEM ENGINEER INTERN, ADVISOR: DR. SACHIN ADNAN Jan. 2014 - Jul. 2014 • Designed a pipeline parallelism based T2 data processing tool on distributed servers, built a T2 data database for statistic analysis of storage system’s workload, and generated a series of reports about read/write characteristics of the storage systems. • Designed a Disk QOS simulation tool, simulated several Disk QOS algorithms and generated a report about their latency and throughput performance. • Learned the configuration steps about VMAX and maintained several servers and storage systems in the lab. Shanghai Academy of Spaceflight Technology RESEARCH INTERN, ADVISOR: PROF. PING CHEN • • • • Shanghai, China Feb. 2012 - May. 2012 Extensively studied a series of independent component analysis algorithms, such as Infomax, FastICA, and JADE. d Designed a simulation tool for satellite communication in Matlab. Designed a satellite signal reception algorithm based on blind signal separation and independent component analysis. Simulated and optimized the performance of this algorithm for real satellite signals. Anjin Medical Technology (Beijing) Co., Ltd Beijing, China EMBEDDED ENGINEER INTERN, ADVISOR: DR. WEI ZHAO Nov. 2010 - Apr. 2011 • Developed the circuit for the LCD module of the surgery workstation. • Designed a GUI in C for the same LCD module. • Learned and supported the development of CPU module of the surgery workstation. Projects Subset-Sum Problem Solvers: Implementation and Evaluation COURSE PROJECT FOR COMBINATORIAL OPTIMIZATION, ADVISOR: PROF. WALEED MELEIS • • • • Boston, Massachusetts Sep. 2013 - Dec. 2013 Developed a benchmark that consists of instances generated from different test problems for the evaluation of Subset-Sum Problem. Looked up all complexity results related to the Subset-sum Problem and its sub-problems. Studied and implemented the exhaustive, greedy, ILP, deepest decent and tabu search algorithms for the Subset-sum Problem. Evaluated the performance of Subset-sum Problem solvers and gave the best solver for different types of Subset-sum Problems. Face Recognition Using Eigenfaces Boston, Massachusetts CODING PROJECT FOR COGNITIVE SYSTEM LABORATORY INTERVIEW, ADVISOR: PROF. DENIZ ERDOGMUS June. 2013 • Wrote a C++ API for Matlab, which allows the C++ ALGIB package to communicate with Matlab. • Studied and implemented a fast PCA algorithm which proved to have a better performance on face recognition. • Implemented the eigenface method based on the fast PCA algorithm and C++ ALGIB package, and classified human face images with the AT&T face database. MARCH 14, 2016 HAOFU LIAO · CURRICULUM VITAE 2 Pedestrian Detection with HOG Descriptors and SVM COURSE PROJECT FOR PATTERN RECOGNITION, ADVISOR: PROF. JENNIFER DY Boston, Massachusetts Jan. 2013 - Apr. 2013 • Implemented several static HOG descriptors in Matlab. • Designed and trained a SVM classifier for HOG descriptors based on INRIA Person dataset and MIT pedestrian dataset. • Developed a human detection toolbox with GUI supported in Matlab. Spatial Domain Image Enhancement Toolbox Using Matlab COURSE PROJECT FOR TWO DIMENSIONAL IMAGE & SIGNAL PROCESSING, ADVISOR: PROF. VINAY INGLE Boston, Massachusetts Sep. 2012 - Dec. 2012 • Studied and implemented a set of spatial domain image enhancement techniques in Matlab. • Built an image enhancement toolbox to demonstrate the relationship between transformation function and the processing results. • Analyzed effectiveness of different image enhancement techniques. Cloud based Intelligent Irrigation Systems Beijing, China GROUP LEARDER, BUPT INNOVATION LABORATORY, ADVISOR: PROF. LI GUO Jan. 2011 - Mar. 2011 • Established the develop environment (OS and interfaces) for embedded irrigation device. • Created a database for storing historical irrigation data such as temperature and humidity on Windows Azure cloud platform. • Developed a Windows Communication Foundation (WFC) based backend communication tool, which supported web communication between embedded irrigation device and smartphone or internet users. Publication Feb. 2016 Haofu Liao, Yucheng Li, Tianran Hu, Jiebo Luo, “Inferring Restaurant Styles by Mining Crowd Sourced Photos from User-Review Websites,” submitted to Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2016. Nov. 2012 Ping Chen, Xinghua Dong, Huichao Zhou, “Communication System Simulation Experiments Tutorial Based on Scilab\SCICOS”, Tingshua University Press, Nov. 16, 2012. Contributed Section 3.1, 3.3 and 3.4 on SCICOS development introduction. Contributed Section 3.1, 3.3 and 3.4 on SCICOS development introduction. Oct. 2011 Haofu Liao, Ping Chen, “Explore a Way of Improving the Computational Efficiency of the Block in SCILAB,” in Proceedings of 2011 IEEE International Workshop on OSSC, pp. 13-16, Oct. 12-14, 2011. Oct. 2011 Zhenzi Liu, Ping Chen, Haofu Liao, Huichao Zhou, “the Modeling and Simulation of Rayleigh Flat Fading Channels based on Scilab”, in Proceedings of 2011 IEEE International Workshop on OSSC, pp. 27-31, Oct. 12-14, 2011. Skills Programming Python, C/C++, Matlab, Bash Script, C#, Perl, LaTeX Parallel Programming Pthreads, CUDA, C++ Multithreading Library, OpenMP, .NET Framework for Parallel Programming, Hadoop Other MongoDB, Git, Makefile, SQL Server, Apache Tomcat, Wordpress MARCH 14, 2016 HAOFU LIAO · CURRICULUM VITAE 3