Home

Yucheng Wang is Master’s student at UC San Diego in the Computer Science and Engineering department, advised by Prof. Chung-Kuan Cheng. Yucheng’s research interests include graph algorithms and machine learning and optimization and VLSI layout. His main focus is on VLSI placement problem. Prior to pursing a Master’s degree, Yucheng has also worked on deep learning models on AFib detection.

SKILLS

• Language: Mandarin (native), English (fluent)
• Computer Science:
  • Language: C, C++, Java, Python, SQL, Bash Script, Assembly (Arm, X86)
  • Packages: JAX, Pytorch, Tensorflow, Pandas, Numpy

EDUCATION BACKGROUND

University of California San Diego

M.S. in Computer Science       Sep. 2021 – May. 2023

• Major GPA:3.66/4.00
• Relevant Coursework: Convex Optimization, Computer Vision, Natural Language Processing, Operating System

Purdue University

B.S. in Computer Science       Jan. 2019 – May. 2021

• Cumulative GPA: 3.94/4.00, Major GPA:3.93/4.00
• Relevant Coursework: Analysis Algorithm, Data Mining, Machine Learning, Image Processing, Data Structure, Computer Architecture, Discrete Mathematics, Database, Linear Algebra, Probability
• Honor: Dean’s List & Semester Honors 2019-2021

RESEARCH EXPERIENCE

• Graphical Research: Eigenvalue Problem with Linear Constraints       Research Assistant, Advised by CK Cheng, Distinguished Professor
  • Formulated an initialization method for analytical placement of VLSI designs to minimize wirelength and density.
  • Derived an efficient algorithm from Spectrum theorem and QCQP that generate initial layout for millions of cells.
• Additive Feedforward Neural Network on Afib Detection Research       Research Assistant, Advised by Xiaoqian Wang, Assistant Professor
  • Designed and implemented Additive Feedforward Learning algorithm on multiple machine learning models to detect heart failure.
  • Performed data cleaning and relabeling on ECG dataset from 2017 PhysioNet/CinC Challenge to be compatible with the additive layers’ setting for further evaluation
  • Used Numpy and Pandas packages to build Logistic Classifiers from scratch to integrate the additive layers, achieving an accuracy over 80% on heart failure detection
• Application of AI and Big Data in Quantitative Transaction       Research Participant
  • Collected and preprocessed CSI 300 Index from 1996 to 2018, and applied factor analysis to extract useful factors like idiosyncratic risk, firm size, asset to market ratio, dividend to price ratio, etc.
  • Applied Machine Learning models like XGBoost, SVM and KNN to compare with base model, Linear regression
  • Applied Fama-French three-factor and five-factor model to evaluate the Machine Learning models

ANDROID PROJECTS

• SchoolBit       Independent Android Development
  • Conducted a comprehensive feature update on Blackboard, an outdated educational platform, to enhance user experiences.
  • Designed and implemented UI assets and individual performance interfaces, allowing students to look at their own scores.
  • Implemented the app from scratch with a user-friendly interface and performance rating system for each course.
• Questa       Independent Android Development
  • Used Java and Android Studio to create a task management app that based on RPG quest mechanics with reward system
  • Designed and painted the entire UI and all pixel assets with Adobe XD and then replicated the design to Android Studio
  • Published on Google Play on Oct.7th