Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

blogs

experience

➣ IT Club President

Shandong Experimental High School, China, 2017-2019

  • One of the Executive Board and Manager Board members in School IT Club

➣ Teaching Assistant

Mathematics, Statistics and Computer Science Department, Macalester College, Spring 2020 - PRESENT 2020

  • Teaching Assistant(Preceptor) in COMP 394 Reinforcement Learning, MATH 378 Complex Analysis, COMP 128 Data Structure, STAT/COMP 112 Intro to the Data Science at Macalester College

➣ Computer Vision Research Intern

Research and Development Department, Thorough Images, Spring 2021

  • Full time research internship for learning, understanding, developing, and optimizing computer vision models to complete intelligent solutions for pathological images at Thorough Images.

honors

projects

➣ Design and Optimization of Comb Drive Accelerator for High Frequency Oscillation

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspCOMSOL</p>

This project is mentored by Albert Chen at UCLA and the final paper was published.
In this work, using COMSOL and MATLAB, a finite element code was used for the design, optimization and visualization of a comb drive accelerator.


</article> </div>

➣ Mathematical Modeling for Drone Light Show

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspMATLAB</p>

This is the final paper for the 20th Annual High School Mathematical Contest in Modeling (HiMCM), problem A.
Using MATLAB, I built models to determine the required number of drones and every drone’s initial location during the process. Any simple pictures or icons could be designed for the drone show by my MATLAB program.View PDF


</article> </div>

➣ House Price Prediction

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspPython</p>

This is the first practice for machine learning and for Kaggle competition: House Prices: Advanced Regression Techniques. Using Ridge, Lasso, LGBM, XGB, Stacking CV Regressor, and etc, to reach Score(mean absolute error): 11977.59807; 13th place out of 19,465 teams (0.06%) For more information, please see my project directly or visit my Github Repository.


</article> </div>

➣ Area Coverage with Unmanned Aerial Vehicles Using Reinforcement Learning

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspPython</p>

2020 Summer Research with professor Esra Kadioglu Urtis, and students Aaron Gould, Elisabeth Landgren, and Fan Zhang at Macalester College. We developed and implemented reinforcement learning based algorithms with simulations and a gym environment for coverage path planning for multiple drones using Actor Critic using Kronecker-Factored Trust Region (ACKTR) deep reinforcement learning method provided by Stable Baselines(OpenAI) in Python(My Reseaerch Report, My Github Repository).


</article> </div>

➣ The Power of Topological Data Analysis for Machine Learning

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspPython</p>

This is the project for the capstone course MATH 471 Topology at Macalester College, discussing the power of topological data analysis for machine learning. Due to the time limit, the test for methodology was only on the MNIST dataset. Digital Artifact Notebook, GitHub Repo


</article> </div>

➣ IHC HER-2 Autonomous Scoring

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbspPython</p>

Full time research internship for learning, understanding, developing, and optimizing computer vision models to complete intelligent solutions for pathological images at Thorough Images.(editing…)


</article> </div>

publications

➣ Accurate Dynamic SLAM using CRF-based Long-term Consistency

Zheng-Jun Du, Shi-Sheng Huang, Tai-Jiang Mu, Qunhe Zhao, Ralph Martin, Kun Xu.

IEEE Transactions on Visualization and Computer Graphics (TVCG) 2020

We propose a novel RGB-D SLAM approach for accurate camera pose tracking in dynamic environments. We provide a more accurate dynamic 3D landmark detection method, followed by the use of long-term consistency via conditional random fields. Evaluation on the TUM and Bonn RGB-D dynamic datasets shows that their approach achieves state-of-the-art accuracy.

PAPER   

➣ Video Recoloring via Spatial-Temporal Geometric Palettes

Zheng-Jun Du, Kai-Xiang Lei, Kun Xu, Jianchao Tan, Yotam Gingold

ACM SIGGRAPH 2021

We introduce a spatial-temporal geometry-based approach to video recoloring. Specifically, its core is a 4D skew polytope with a few vertices that approximately encloses the video pixels in color and time, which implicitly defines time-varying palettes through slicing of the 4D skew polytope at specific time values.

PAPER   

rl

➣ Chapter 2 Multi-armed Bandits

16 minute read

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</p>

“Reinforcement Learning: An Introduction” by Richard Sutton & Andrew Barto, 2nd Ed


</article> </div>

➣ Chapter 3 Finite Markov Decision Processes

34 minute read

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</p>

“Reinforcement Learning: An Introduction” by Richard Sutton & Andrew Barto, 2nd Ed


</article> </div>

➣ Chapter 4 Dynamics Programming

25 minute read

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</p>

“Reinforcement Learning: An Introduction” by Richard Sutton & Andrew Barto, 2nd Ed


</article> </div>

➣ Chapter 5 Monte Carlo Methods

36 minute read

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</p>

“Reinforcement Learning: An Introduction” by Richard Sutton & Andrew Barto, 2nd Ed


</article> </div>

➣ Chapter 6 Temporal-Difference Learning

61 minute read

&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</p>

“Reinforcement Learning: An Introduction” by Richard Sutton & Andrew Barto, 2nd Ed


</article> </div>