Hang Yang

Hang Yang

he/him

PhD candidate in Mathematics at UMass Amherst. Researching machine learning, robotics, and deep learning — including mobile motion capture with robot dogs, invertible neural networks for image steganography, and vision transformers for medical AI.

Hang Yang

I am a PhD candidate in the Department of Mathematics and Statistics at the University of Massachusetts Amherst. My research lies at the intersection of machine learning, applied mathematics, and robotics, with a focus on deep learning for image steganography, vision transformers for medical image analysis, and mobile motion capture with legged robots.

Education

  • Doctor of Science in Mathematics — University of Massachusetts Amherst (UMass), Sep 2024–Present · GPA: 3.95/4.0
  • Master of Science in Mathematics — China Agricultural University (CAU), Sep 2021–Jul 2024 · GPA: 3.80/4.0 (WES: 3.86/4.0)
  • Bachelor of Science in Mathematics — China Agricultural University (CAU), Sep 2017–Jul 2021 · GPA: 3.65/4.0 (WES: 3.61/4.0)

Research Interests

  • Mobile Motion Capture — using a robot dog as a mobile motion capture platform to enable flexible, large-scale human motion recording beyond the constraints of traditional fixed studio setups
  • Robotics — legged robot perception, autonomy, and human-robot interaction
  • Image Steganography — developing invertible neural networks that hide secret images within carrier images robustly and securely (PRIS, DKiS)
  • Medical AI — applying transformer-based architectures to automated disease detection from medical imaging (CovidViT)
  • Invertible Neural Networks — designing networks with exact invertibility for lossless information recovery

Publications

  • DKiS: Decay weight invertible image steganography with private key — Neural Networks (2025) · DOI · GitHub
  • PRIS: Practical robust invertible network for image steganography — Engineering Applications of Artificial Intelligence (2024) · DOI · GitHub
  • CovidViT: A novel neural network with self-attention mechanism to detect Covid-19 through X-ray images — International Journal of Machine Learning and Cybernetics (2023) · DOI

Other Projects

  • Personal Website — built yanghang.site using Vue.js and Django
  • DQN with N-Steps — reproduced the DQN algorithm from Human-Level Control through Deep Reinforcement Learning and introduced an n-step return method for improved performance
  • Game AI — designed and implemented automatic control for Airplane Wars and Tic-Tac-Toe using Policy Gradient and Q-table methods
  • Automatic Figure Generation — implemented a GAN trained on the MNIST dataset for automatic figure generation
  • Neural Network Pathfinding — implemented automatic pathfinding based on neural networks

Contact