Dr. Brent Lessley

Ph.D. Graduate, February 2019
CIS Department
University of Oregon

Area Exam: [PDF]
Dissertation: [PDF]
Google Scholar page

Bio

Brent is a Ph.D. alumnus of the group, defending his thesis in February 2019. Upon graduation, he accepted a full-time position at Verb Surgical, a joint venture between Google and Johnson & Johnson that is located on the Google campus in Mountain View, CA.

During his Ph.D. research, Brent explored many-core parallel solutions for graph and scientific visualization algorithms, including maximal clique enumeration, image segmentation, and external facelist calculation. These works led him to study the optimal patterns for index-based search for scientific visualization and analysis workloads for his dissertation question. Further, Brent's work on external facelist calculation led to a new idea for hashing on many-core platforms, HashFight, which Brent pursued in work that complemented his thesis question. He spent the summers of 2016 and 2017 at Lawrence Berkeley National Laboratory working with Dr. Wes Bethel.

Prior to joining the CDUX research group, Brent studied data mining and machine learning, with a particular focus on adversarial (robust) machine learning and spam detection. He received a B.A. in Business Administration (2009) and M.S. in Computer Science (2011), both from the University of Washington. Hobbies include singing, running, and learning languages.

Publications


Data-Parallel Hashing Techniques for GPU Architectures
Brenton Lessley and Hank Childs
IEEE Transactions on Parallel and Distributed Systems, January 2020

[LINK]     [BIB]

HashFight: A Platform-Portable Hash Table for Multi-Core and Many-Core Architectures
Brenton Lessley, Shaomeng Li, and Hank Childs
IS&T International Symposium on Electronic Imaging: Visualization and Data Analysis (VDA), Burlingame, CA, January 2020

[PDF]     [BIB]

DPP-PMRF: Rethinking Optimization for a Probabilistic Graphical Model Using Data-Parallel Primitives
Brenton Lessley, Talita Perciano, Colleen Heinemann, David Camp, Hank Childs, and E. Wes Bethel
IEEE Symposium on Large Data Analysis and Visualization (LDAV), Berlin, Germany, October 2018

[PDF]     [BIB]

Maximal Clique Enumeration with Data-Parallel Primitives
Brent Lessley, Talita Perciano, Manish Mathai, Hank Childs, and E. Wes Bethel
IEEE Symposium on Large Data Analysis and Visualization (LDAV), Phoenix, AZ, October 2017

[PDF]     [BIB]

Techniques for Data-Parallel Searching for Duplicate Elements
Brent Lessley, Kenneth Moreland, Matt Larsen, and Hank Childs
IEEE Symposium on Large Data Analysis and Visualization (LDAV), Phoenix, AZ, October 2017

[PDF]     [BIB]

External Facelist Calculation with Data-Parallel Primitives
Brent Lessley, Roba Binyahib, Robert Maynard, and Hank Childs
EuroGraphics Symposium on Parallel Graphics and Visualization (EGPGV), Groningen, The Netherlands, June 2016

[PDF]     [BIB]