Ziqiao Ma

Here is a virtual but warm greeting from Ziqiao, as you just came across his page! Well, many of my non-Chinese friends found it hard to pronounce my Chinese name, so...it's absolutely fine to just call me Martin alternatively.

I am a Senior student at the University of Michigan in Computer Science , with a dual degree in Electrical and Computer Engineering at Shanghai Jiao Tong University.

I have a passionate interest in machine learning, especially machine learning on graph-structured data. At the moment, I am intrigued by the studies of active learning, interpretability, and robustness on Graph Neural Network. I am also actively exploring how state-of-the-art machine learning breakthroughs can play a role in enabled interdisciplines.

News

• [Temporary Update] Transcript for Fall 2020 is updated here for graduate school committee visiting this page.
• [Dec. 2020] I will be the Instructional Aide for EECS 492 Winter 2021, working with Professor Chai!
• [Aug. 2020] I moved my blogs here. Let me know what you want to see!

Experience

Education

• University of Michigan, Ann Arbor
  B.S. in Computer Science || GPA: 4.00/4.00 || Sept. 2019 - Present
  
  • Natural Language Processing (A+)
  • Machine Learning (A+)
  • Artificial Intelligence (A)
  • Data Structure and Algorithms (A)

  • Deep Learning for CV (A+)
  • Computer Vision (A+)
  • Numerical Analysis (A+)
  • Matrix Theory (A+)
• Shanghai Jiao Tong University
  B.S. in Electrical and Computer Engineering || GPA: 3.81/4.00 || Sept. 2017 - Present
  
  • Programming and Data Structure (A)
  • Logic Design (A)

  • Discrete Mathematics (A)
  • Honors Mathematics II-IV (A-, A, A-)
• Technische Universität Berlin
  Winter Program || Jan. 2019 - Feb. 2019
  
  • Programming in Java (A)
• Chengdu Foreign Languages School
  High School || Sept. 2011 - June 2017

Timeline

Selected Projects

Research

GNN Prediction Interpretation: Spatial Cell Patterns of T2D Islets

[report: upon request][code]

• Instructed by Professor Jie Liu and Marcela Brissova.
• Redesign the GNN Explainer to perform label-wise explanation, validate the model on T2D islet samples and extract predictive spatial patterns.

Active learning on Graph Neural Network via Hierarchical Graph Partitioning

[code: upon request]

• Instructed by Professor Qiaozhu Mei.
• A practical active learning query strategy on Graph Neural Networks.

Correlational Information Utilization in Spatial Temporal GCN on Traffic Data

[code: synthetic][code: validation]

• Instructed by Professor Qiaozhu Mei.
• Confirm the incapability of GNNs to capture correlational information by synthetic simulation.
• Validate lower training loss of ST‐GCN on traffic data with gaussian copula loss.

Exception Handling in Autonomous Vehicles via Human Language Communication

[slides]

• Instructed by Professor Joyce Chai.
• Develop a smart interface that takes human interaction by language instructions and handle the exceptions in autonomous vehicles.

Courses

Towards More Knowledgeable Reasoning: Experiments on Natural Language Inference

[report][code]

• Final project of EECS 595: Natural Language Processing. Instructed by Professor Joyce Chai.
• Exploited external knowledge to perform NLI tasks including Question Answering, Conversation Entailment and Plausible Inference.
• Experimented on different finetuning strategies and utilized knowledge graphs for GCN-based reasoning.

Scalable and Diverse Neural Style Transfer with Generative Adversarial Networks

[report][slides][code]

• Final project of EECS 442: Computer Vision. Instructed by Professor Andrew Owens.
• Reproduced CycleGAN, StarGAN v1 and v2 for unpaired image translation.
• Evaluated them quantitatively on neural style transfer tasks with FID score.

Instagram MINI

[description][code: upon request]

• Project of EECS485: Web Systems. Instructed by Professor Andrew DeOrio
• A cloned client-side dynamic Instagram.

Glove Band

[report]

• Final Project of VG100, Intro to Engineering. Instructed by Professor Yanfeng Shen.
• A pair of multi-instrumental MIDI-controlling gloves.

• Full Portfolio

Misc

• I was born and raised up in Chengdu, the home of pandas.
• I am proud of Chengdu Foreign Languages School, my high school, and identify myself as a CFLSer. 成外人永远不会成外人。
• My dream school during high school was The University of Hong Kong (HKU), yet I declined the offer of BSc(QFin) in pursuit of my engineering dream.
• I love literature and drama. I am particularly interested in Shakespeare's plays, ancient Chinese drama and modern Asian literature. Here is one of my essays about Cross-dressed Heroines in Comedies.
• I enjoyed travelling and have been to 12 countries. Maybe I have been to your hometown?
• I am a super fan of Nightwish. My favourite album is Imaginaerum.

Mental Support

• I once suffered from serious depression but I am lucky enough to have recovered. If you need any mental support, please feel free to send me an email. Sometimes it is easier to share your feelings to a stranger.

Get In Touch

You are welcome to drop me a message :)