Daniel Lowd

Portrait of Daniel Lowd

E-mail: lowd at cs dot uoregon dot edu
Office: 262 Deschutes Hall
Phone: 541-346-4154
Fax: 541-346-5373

Mailing Address:
Department of Computer and Information Science
University of Oregon, Eugene, OR 97403

I am an Assistant Professor in the Department of Computer and Information Science at the University of Oregon.

My research interests include learning and inference with probabilistic graphical models, adversarial machine learning, and statistical relational machine learning.

I also maintain Libra, an open-source toolkit for Learning and Inference in Bayesian networks, Random fields, and Arithmetic circuits.
Latest version: 1.1.1, released on 3/28/2015.


October 2015

Our paper A Probabilistic Approach to Knowledge Translation (with Shangpu Jiang and Dejing Dou) has been accepted to AAAI 2016

September 2015

Our paper Ontology Matching with Knowledge Rules (with Shangpu Jiang and Dejing Dou) was published in DEXA 2015 and won the Best Paper Award!

July 2015

Automated Attacks on Compression-Based Classifiers was accepted to the ACM AISec 2015 workshop (with Igor Burago).

June 2015

I received an ARO Young Investigator Award for my proposal on "Inferring Trustworthiness and Deceit in Adversarial Relational Models"!

May 2015

I will (again) be serving as Proceedings Chair for the Conference on Uncertainty in Artificial Intelligence (UAI 2015) and Workshops Cochair for AAAI 2016.

March 2015

Libra version 1.1.1 is now available! Includes better documentation, better interface, bug fixes, and more!

September 2014

New NSF grant funded -- "EAGER: Machine Learning to Combat Adversarial Attacks"!

June 2014

Presented two papers at ICML 2014 - On Robustness and Regularization of Structural Support Vector Machines (with Ali Torkamani) and Learning Sum-Product Networks with Direct and Indirect Interactions (with Pedram Rooshenas)

...More News...