Shangyi Guo 郭 尚仪
Hello, I’m Shawn, currently a graduate student pursuing a Master’s in Applied Urban Science and Informatics at New York University Center for Urban Science + Progress (NYU CUSP). My academic background combines Computer Science and the Internet of Things (IoT), which I studied at Beijing University of Technology. In addition, I bring practical experience across deep learning, data analysis, and environmental studies, making me passionate about leveraging data-driven solutions to address urban challenges to support environmental justice and resilience.
Currently, I’m honored to serve as a research assistant in Dr. Yi Yin’s lab at NYU, where I focus on using deep learning methods to leverage “data-driven prople-centered decision-making”. Before this, I worked on various projects in the AI field, including improving pollutant emission forecasting using time series Transformer models for the EU’s Horizon 2020 AQ-WATCH project in Institute of Urban Safety and Environmental Science,Beijing Academy of Science and Technology; improving interpretability of diagnosis of medical image like CT scan with Causal Representation Learning methods; bilding and deploying deep learning models for sleep staging and anoamly warning for Non-contact human vital signs monitoring system with UWB radar(3rd price in HUAWEI ICT Competition national level).
Prior to joining NYU, I worked as the CTO for an stealth AI-music startup, where I focused on system integration and deep learningYinLanMusic. Additionally, I’ve had the opportunity to intern with Microsoft Research Asia in Beijing, handling public communications and content creation for AI-driven research projects.
Research focus
Yeah the experience I listed above might seems a bit messy, but if we take a closer look, it acctually have an underlying focus, which is solving the big problems of today’s AI applications to achieve: interpretable, resoponsible and reliable AI.
Since current AI models could not take responsibility for their decisions, we cannot fully automate processes without human supervision in fields such as healthcare and education, where decisions can have irreversible effects on individuals. Furthermore, the lack of explainability in these models prevents us from determining whether the output results can be trusted.
That is why I am actively learning more about Causal AI based on causal inference.
Timeline
- Institute of Urban Safety and Environmental Science,Beijing Academy of Science and Technology.(ongoing 2023.7-2023.8)
- Microsoft Research Asia MCAPS ARD communication intern.(finished 2023.3-2023.5)
- HUAWEI ICT Competition innovation track China region third prize(finished 2023.4)