Christopher Jermaine, professor of computer science, and his group won the best paper runner up award at the 45th International Conference on Very Large Databases (VLDB).
He shared the award with Ph.D. students Dimitrije Jankov, Shangyu Luo, Binhang Yuan, Jacob Gao (Ph.D. ’18) and Jia Zou, assistant professor of computer science and engineering at Arizona State University.
“Declarative Recursive Computation of an RDBMS, or, Why You Should Use a Database For Distributed Machine Learning” is the title of the paper. The group competed with 677 research papers submitted for presentation at the conference.
The paper explains how to make a very small set of changes to a modern relational database management system (RDBMS) to make it suitable for distributed learning computations. Jermaine said his group plans to study the use of a database system as an ML platform.
“While ML platforms such as TensorFlow and PyTorch are very popular, they can be difficult to use for distributed machine learning or for building very large models. Our paper showed that building such a system on top of database technology can alleviate such problems,” he said.
VLDB took place in Los Angeles, California August 26 – 30, 2019.
Cintia Listenbee, Communications and Marketing Specialist at Rice University