I am a Professor of Computer Science at the University of Oregon. I work on adversarial machine learning, tractable probabilistic modeling, and statistical relational learning.
CV · Publications · Research
Email lowd@uoregon.edu
Selected Publications
- Reducing Certified Regression to Certified Classification for General Poisoning Attacks
Zayd Hammoudeh and Daniel Lowd — SaTML 2023 (2023)
Certified defenses that reduce robust regression to classification, expanding guarantees against poisoning attacks. - Instance-Based Uncertainty Estimation for Gradient-Boosted Regression Trees
Jonathan Brophy and Daniel Lowd — NeurIPS 2022 (2022)
Calibrated uncertainty estimates for tree ensembles using local influence of similar training examples. - Machine Unlearning for Random Forests
Jonathan Brophy and Daniel Lowd — ICML 2021 (2021)
Algorithms that efficiently remove the effect of individual samples from trained random forests.
You can also find recent publications on Google Scholar or Semantic Scholar, which may be more up to date. See the full Publications page for the complete list.
Note to Prospective Students
If you’re interested in collaborating or joining the group, please read the prospective student FAQ first.