Facebook open-sources deep learning recommendation model DLRM

Machine learning in production environments is on the rise. Facebook is now joining the list of companies open-sourcing their technology to improve AI applications in personalisation and recommendation models.


Facebook has disclosed the open source release of Deep Learning Recommendation Model (DLRM), an advanced AI model for providing personalised results in production environments.

The social media giant made the announcement through a blog post on their Facebook AI site on Tuesday.

In it, Facebook research scientists Maxim Naumov and Dheevatsa Mudigere highlighted the importance of deep learning in personalisation and recommendation models.

They explained how these models differ substantially from other deep learning models because they must be able to work with categorical data, which is used to describe higher-level attributes.

“It can be challenging for a neural network to work efficiently with this kind of sparse data, and the lack of publicly available details of representative models and data sets has slowed the research community’s progress”, the scientists said.

Together with the new model, Facebook also released an accompanying paper aiming to “help the community find new ways to address the unique challenges presented by this class of models.”

Facebook said that, by making DLRM open-source, they also hope to encourage further algorithmic experimentation, modelling, system co-design, and benchmarking of recommendation engines.

“This in turn will lead to new models and more efficient systems that can provide more relevant content to people using a wide range of digital services.”

DLRM can be downloaded from GitHub, and implementations of the model are also available for Facebook’s PyTorch and Caffe2, and Glow C++.

Recent Articles