Welcome, my name is Zexi Huang. I am a PhD student at the Department of Computer Science, University of California, Santa Barbara (UCSB), starting from Fall 2018. Before joining UCSB, I received my Bachelor's Degree in Computer Science and Technology with the highest honor at Yingcai Honors College (elite school of top 5% undergraduates), University of Electronic Science and Technology of China (UESTC). During my undergraduate years, I spent half a year as a visiting research assistant at the School of Computer Science and Engineering, Nanyang Technological University (NTU), and enjoyed a nice semester as an exchange student at the College of Electrical and Computer Engineering, National Chiao Tung University.
My research interests span the analysis of social, informational, and biological networks with machine learning and data mining techniques. At UCSB, I work in Dynamic Networks: Analysis and Modeling (Dynamo) Lab, with Prof. Ambuj Singh as my advisor. My current research topic is network embedding and dynamics. Previously, I have interned at Books Tech, Amazon, as an Applied Scientist working on graph-based fraud detection in Kindle Direct Publishing. I have also interned at Computational Intelligence Lab, NTU on transfer learning framework for community detection in multiplex networks. As an undergraduate, I was a research assistant at Data Mining Lab, Big Data Research Center, UESTC, with Prof. Junming Shao as my advisor. I worked on dynamics-based algorithms for detecting overlapping communities at that time.
Sep. 2018 - Exp. Jun. 2023
- Course GPA: 4.0/4.0.
- Advisor: Prof. Ambuj Singh. Research topics: Network Science, Machine Learning, Data Mining.
Sep. 2014 - Jun. 2018
- GPA: 3.96/4.0, Avg. Score: 92.79/100 (1/90 in freshman year, 1/87 in sophomore year, 1/93 in junior year).
- Advisors: Prof. Junming Shao and Prof. Sinno Jialin Pan. Thesis: Transfer Learning for Community Detection.
Feb. 2016 - Jun. 2016
- GPA:4.0/4.0, Avg. Score: 96.44/100, Straight A+.
Jun. 2020 - Sep. 2020
- Proposed to augment existing fraud detection methods with graph-based machine learning models in Kindle Direct Publishing.
- Designed and implemented various heuristics and an embedding framework for attributed heterogeneous multiplex networks.
- The models are deployed to graphs with millions of nodes and billions of edges in production. Validation results show that they can surface up to 15 times more fraud compared to the existing processes.
Sep. 2017 - Feb. 2018
- Proposed to refine community detection results in some layers with transferred knowledge from other layers in multiplex networks.
- Designed a representation-based community detection framework and implemented it with an extended symmetric NMF approach.
- Our algorithm outperforms other representation-based community detection algorithms, especially when the target layer is noisy.
Sep. 2018 - Oct. 2020
- Paper submitted to WWW’21. Details omitted for double-blind review.
Jan. 2020 - Aug. 2020
- Extended the Prospect Theory to model group-level risky decision making dynamics with an interpersonal influence system.
- Results on two human-subject experiments show that the group behavior shifts towards consensus and is explained by the influence.
Jul. 2016 - Aug. 2017
- Proposed an intuitive algorithm for fast overlapping community detection in networks.
- Introduced gain and loss functions from game theory to model the intrinsic dynamics between nodes.
- Results are comparable to state-of-the-art algorithms.
2020 - 2021
- Python, Matlab, Java, C/C++, SQL, Cypher, LaTeX, Verilog, ARM
- Chinese (native), English (fluent), German (basic)
- General Judge of SB Hacks V Hackathon
- Student Speaker of UESTC Open House for Prospective Students
- Senior Student Speaker of Yingcai Honors College Opening Ceremony
- Trainee of Innovation Technologies Training Course at HKUST
- Member of Academic Group of Tang Lixin Elite Student Club of UESTC
- Ambassador of Cross-culture Communication Group of UESTC