معرفی
Brian Hutchinson is a Professor in the Computer Science Department at Western Washington University (WWU) and holds a joint appointment as Staff Scientist at the Pacific Northwest National Laboratory (PNNL). His research focuses on applying machine learning and deep learning to interdisciplinary problems in astronomy, biology, climate science, and education. He leads the Hutchinson Machine Learning Research group (hutchresear.ch), which collaborates with domain experts across multiple disciplines.
- Affiliations: WWU (since 2013), PNNL (since 2017), University of Washington (PhD 2013), Microsoft Research (internships 2010–2011)
- Educations: PhD in Electrical Engineering (UW 2013), MS in EE (UW 2009), MS/BS in CS and BA in Linguistics (WWU 2005–2006)
His research interests include adversarial learning, multimodal learning, spoken language processing, and applications in astrophysics (e.g., star classification), bioinformatics (e.g., protein mutation analysis), and climate modeling (e.g., Earth System Model emulation). His work bridges theory and practice, with contributions to neural network architectures like Tensor Deep Stacking Networks and frameworks such as RoseNet and Gaia Net.
Recent publications emphasize machine learning for climate science (e.g., DiffESM for precipitation emulation), astronomy (e.g., APOGEE Net for stellar spectral analysis), and educational tools (e.g., machine learning in chemistry education). He has also explored cybersecurity applications, defect detection in materials science, and classroom activity detection via audio analysis.
- Awards: WWU CS Professor of the Year (2015, 2023), WWU Faculty Mentor of the Year (2020), UW EE Outstanding Research Award (2013).
- Grants/Advising: Leads projects in climate modeling, astrophysics, and energy efficiency. Supervised numerous students in interdisciplinary collaborations.
His lab develops open-source tools and collaborates with national labs (e.g., PNNL) and international surveys (e.g., Sloan Digital Sky Survey). Current work includes advancing diffusion models for climate data emulation and improving interpretability in astrophysical machine learning systems.
