Ruiyi Fang (方睿怡)

Hi! This is Ruiyi (Raelyn)! I am a first-year doctor student at Yamakawa Lab in UTokyo, advised by Prof. Yawakama.

Previously, I received my master's degree in Artificial Intelligence from Central South University, fortunately advised by Prof. Kai Wang. I received B.E. from Chang'an University.

I am broadly interested in AI for robotics, embodied AI, Vision-Language-Action(VLA) models, and robot learning. Specifically, my work focuses on imitation learning based robot manipulation, especially in few-demonstration scenarios. I enjoy thinking about how to enable robots to acquire complex skills from limited data, enhancing autonomy and adaptability in real-world applications.

Please feel free to contact me with email!

 ~  Email  /  CV  /  Github /  Google Scholar ~ 

News

Oct'25

Research

In this study, we introduce a dynamic domain adversarial network (DDAN) for dynamic few-sample fault detection in industry. To tackle the dynamic characteristic in industrial data, a special multi-step difference module incorporating the self-attention mechanism is designed. Moreover, considering the few-sample problem (or cold-start problem), the DDAN is trained in an adversarial framework to transfer the domain-invariant features.

In this work, we introduce an adversarial-based unsupervised network for fault detection in few-sample scenarios. Due to the significant disparity in data quantity between the target domain and the source domain, the adversarial training process inevitably encounters an imbalance problem. To address this issue, we propose a domain imbalance aware margin (DIAM) loss to rebalance the domain margin between the two domains.

We propose a method called WDAA, which utilizes an adversarial framework to transfer knowledge from a data-rich source domain to a data-poor target domain. To further enhance the domain adaptation performance, we develop a Wasserstein adaptor. The model was validated using data from an industrial process, the Continuous Stirred-Tank Reactor (CSTR).

Education

Awards:

• Outstanding Graduate | 2025
• National Scholarship | 2024
• Post-graduate First-Class Scholarship | 2022 & 2023 & 2024

Awards:

• Academic Excellence Scholarship | 2018 & 2021
• Course Excellence Scholarship | 2018 & 2019 & 2020

This template is a modification to Jon Barron's website. Find the source code to my version here. Feel free to clone it for your own use while attributing the original author Jon Barron.