Debora Sujono
deborausujono -at- gmail -dot- com
ABOUT
I am a PhD candidate (who has mostly abandoned my dissertation) in Computer Science at the University of California, Irvine, advised by Prof. Erik Sudderth. Currently, I am interested in community college teaching and supporting equitable access to computer science education for underrepresented and non-traditional students. In 2023-2024, I was a teaching intern at Irvine Valley College through UCI’s California Community College Internship Program (CCCIP). My prior research interest was in statistical machine learning, specifically on the topics of variational inference and probabilistic programming languages. Previously, I worked as a junior data scientist at MassMutual. I received my MS in Computer Science from the University of Massachusetts, Amherst, where I worked with Prof. Daniel Sheldon, and my BA in Computer Science and Linguistics from Hampshire College. Outside of research, I enjoy yoga, rock climbing, and traveling to tropical diving destinations.
PUBLICATIONS
Variational Inference for Soil Biogeochemical Models. Debora Sujono, Hua Wally Xie, Steven Allison, Erik Sudderth. In AI for Science Workshop, International Conference on Machine Learning (ICML), 2022.
A framework for variational inference and data assimilation of soil biogeochemical models using state space approximations and normalizing flows. Preprint. Hua Wally Xie, Debora Sujono, Tom Ryder, Erik Sudderth, Steven Allison. Submitted to Journal of Advances in Modeling Earth Systems (JAMES), 2022.
Marginalized Stochastic Natural Gradients for Black-Box Variational Inference. Geng Ji, Debora Sujono, Erik Sudderth. In International Conference on Machine Learning (ICML), 2021.
Probabilistic Inference with Generating Functions for Animal Populations. Daniel Sheldon, Kevin Winner, Debora Sujono. In Artificial Intelligence and Conservation, 2019.
Learning in Integer Latent Variable Models with Nested Automatic Differentiation. Daniel Sheldon, Kevin Winner, Debora Sujono. In International Conference on Machine Learning (ICML), 2018.
Exact Inference for Integer Latent-Variable Models. Kevin Winner, Debora Sujono, Daniel Sheldon. In International Conference on Machine Learning (ICML), 2017.