a fourth-year computer science Ph.D. candidate at UC Irvine advised by Erik Sudderth. My primary research area is probabilistic machine learning.
I am obsessed with board games, (recent favorites are Spirit Island, Betrayal Legacy and Root) I'm learning to surf, (I am not good at it) and I'm raising a puppy named Claude (Zoom is hard even for the best of us).
Stochastic Processes are a powerful tool for modeling real-world phenomena, from national elections to microscopic diffusion. For these models to be useful, we must combine domain-specific knowledge with the mathematical tools and tricks necessary for efficient inference.
As real-world decision-making become increasingly automated, ethical questions about the fairness of our models are no longer thought experiments. Left unchecked, algorithms can exacerbate disparities between genders, racial groups, or other legally-protected categories of identity.
The complexity of biological systems has historically made both understanding them and simulating them difficult. Recent gains in computational power and the creation of large-scale data sets provide the potential for many new discoveries.
| Scalable & Stable Surrogates for Flexible Classifiers with Fairness Constraints. H Bendekgey, E Sudderth. NeurIPS 2021. Talks and Presentations: Southern California Machine Learning Symposium 2021 |
| Scaling study of diffusion in dynamic crowded spaces. H Bendekgey, G Huber, and D Yllanes. ArXiv Link Talks and Presentations: March Meeting 2020, March Meeting 2022 |
| Clustering Player Strategies from Variable-Length Game Logs in Dominion H Bendekgey. AAAI Workshop on Knowledge Extraction from Games (KEG), 2019. ArXiv Link |
| Consistency and Reproducibility in U.S. House of Representatives Forecasts H Bendekgey. arXiv preprint arXiv:1811.12466, 2018. ArXiv Link |
