University of Oregon
Daniel Lowd is an Assistant Professor in the Department of Computer and Information Science at the University of Oregon. His research covers a range of topics in statistical machine learning, including statistical relational representations, unifying learning and inference, and adversarial machine learning applications (e.g., spam filtering). In 2009, he coauthored book on Markov logic with Pedro Domingos, published by Morgan & Claypool. He is also the recipient of graduate research fellowships from the National Science Foundation and Microsoft Research.
Daniel Lowd's research is in statistical machine learning, with a focus on statistical relational learning and adversarial machine learning. These areas relax the traditional machine learning assumption that training examples are independent and identically distributed, and that future data will be drawn from the same distribution.
Statistical relational learning focuses on the relationships among the examples, such as links among Web pages, friendship in a social network, or protein interactions in bioinformatics. This is done by by combining a relational representation, such as first-order logic, with a statistical representation, such as Markov or Bayesian networks.
Adversarial machine learning deals with domains where an adversary is actively trying to evade detection. For example, spammers often try to disguise their emails by changing or adding words. By modeling the machine learning problem as a game, we hope to automatically build classifiers that are not so easily tricked.