Qiucheng Wu

I am a 1st year Ph.D. student in Computer Science at University of California, Santa Barbara, under supervision of Prof. Shiyu Chang. I received my bachelor and master's degrees from University of Michigan College of Engineering. I also receivied another bachelor degree from and Shanghai Jiao Tong University UM-SJTU Joint Institute.

I am interested in multimodality problems in Artificial Intelligence. Preivous to my Ph.D. study, my interests span across natural language processing, computer vision, reinforcement learning and data mining. Some of my projects and publications related to these fields are demonstrated below.

Currently, I focus on beyond the state-of-the-art image deblurring methods.

Email: qiucheng@ucsb.edu /  CV  / 

profile photo
Education Experience
UC Santa Barbara
1st year Ph.D. student, 2021

University of Michigan
Masters, Computer Science and Engineering, Sep 2019 - May 2021
Bachelor, Computer Science, Sep 2017 - May 2019
GPA: 3.94/4.00
Major Courses: (At least "A")
  • Natrual Language Processing
  • Situated Language Processing for Emboddied Agents
  • Deep Learning for Vision
  • Data Mining & Advanced Data Mining
  • Advanced Artificial Intelligence
  • Introduction to Machine Learning

Shanghai Jiao Tong University
Bachelor, Electircal and Computer Engineering, Sep 2015 - Aug 2019
GPA: 3.67/4.00
Award: Undergraduate Capstone Team Gold Prize (2019), Dean List(2016, 2017)
Publications and Research Projects

Learning Action Translator for Meta Reinforcement Learning on Sparse Rewards Tasks
Yijie Guo, Qiucheng Wu, Honglak Lee
ICML 2021 Workshop
Submitted to NeurIPS 2021
Comparison between Human Attention and Linguistic Justifications on Images
Qiucheng Wu
Natural Language Processing (EECS 595, FA 19) Course Project, full points (10/10)

In this work, we compare human attention and linguistic justificaiton on Visual Question Answering (VQA) problems. We collect the regions that human focuses on in different images, and we also collect the textual answers, representing them in the images. Finally, we compare two regions in the images to study their relations.

[REPORT] [POSTER]

Landmark Recognition in Vision Language Navigation
Qiucheng Wu, Naihao Deng, Yin Lin
Situated Language Processing in Emboddied Agents (EECS 598, WN 20) Course Project

In this work, we consider the visual language navigation (VLN) problem by examine if the agents can recognize and reach the intermediate landmarks in the whole path. This reveals agents' ability to ground intermediate landmarks with visual information all along the way.

[REPORT] (We use the EMNLP template. This is not a submitted paper.)

Graph Two-sample Testing with Node Embeddings
Qiucheng Wu, Yuze Lou, Shucheng Zhong and Jiaxin Wang
Advanced Data Mining (EECS 576, FA 19) Course Project, full points (40/40)

In this work, we propose a novel method to solve graph two- sample testing problem by combining node embedding methods and statistical two-sample test methods. We have evaluated our method with multiple node embedding methods and conduct experiments on both synthetic datasets and real datasets.

[REPORT] [POSTER]

DataSifterText: Partially Synthetic Text Generation for Sensitive Clinical Notes
Nina Zhou, Qiucheng Wu, Zewen Wu, Simeone Marino, and Ivo Dinov
Submitted to Journal of the American Medical Informatics Association, 2020

In this work, we propose the DataSifter-Text to protect privacy of sensitive textual dataset. The DataSifter-Text protects the data by obfuscating identifiable information in the dataset. We analyze the trade-off between obfuscation and preserving utilities of the data, and DataSifter-Text effectively balances between privacy and utility.

DataSifter: Statistical Obfuscation of Electronic Health Records and Other Sensitive Datasets
Marino, S, Zhou, N, Zhao, Yi, Wang, L, Wu Q, and Dinov,
Journal of Statistical Computation and Simulation, 2018

This is the previous work of DataSifter-Text. In this work, we propose the DataSifter to obfuscate sensitivie dataset while preserve the utility for other researches. I discuss and implement the obfuscation function in R-cpp.

Most Interesting Methodological Advances in 4th Annual Symposium Poster Competition in Michigan Institute for Data Science, University of Michigan.

Compressive Big Data Analytics: An ensemble meta-algorithm for high-dimensional multisource datasets
Simeone Marino, Yi Zhao, Nina Zhou, Yiwang Zhou, Arthur W. Toga, Lu Zhao, Yingsi Jian, Yichen Yang, Yehu Chen, Qiucheng Wu, Jessica Wild, Brandon Cummings, Ivo D. Dinov
PLOS One, 2020

We apply the compressive big data analytics (CBDA) to analyze sailent features and key biomarkers of a high-dimensional dataset.

Most Likely Health Impact in 4th Annual Symposium Poster Competition in Michigan Institute for Data Science, University of Michigan.

Teaching Experience
Graduate Student Instructor: Intro to AI (EECS 492)
College of Engineering, University of Michigan, FA 2020 and WN 2021
Leading discussions, design homework, grade exams, help students in office hours.
Instructor Assistant: Applied Honors Calculus (Vv156, equivalent to MATH 156 in umich)
UM-SJTU Joint Institute, Shanghai Jiao Tong University, FA 2016
Leading discussions, design and grade homework/exams, help students in office hours.
Also I was a grader on Data Mining (EECS 476), Deep Learning (EECS 498), Intro to AI (EECS 592), Applied Linear Algebra (MATH 214), Intro to Computer Organization (EECS 370) for at least 1 semester.
Working Experience
Fitly Inc.
Machine Learning Intern
Nov. 2019 - Apr. 2020

In this internship, I was responsible for designing a personalized model to provide more accurate food classification results for users. My code is now deployed in the real product.
Others
Intel Akraino: Edge Cloud Game Architecture
Undergraduate Capstone Team Gold Prize
UM-SJTU Joint Institute, Shanghai Jiao Tong University, August 2019

In this capstone project, we designed a framework to boost web communications between terminal devices and servers by introducing edge servers. We use kubernets to organize edge servers as different containers. The edge servers are responsible for computing and sending information to terminal devices, leading to more efficient performance.


This website is built using the source code from Jon Barron's public academic website.