Research Area

Ranking & Recommendations

Our research aims to build cutting-edge AI that can understand, inspire, and connect people to what's most meaningful to them. We push the state of the art in applied research to learn from large, diverse datasets, and build powerful and flexible frameworks that allow us to train and deploy models at huge scale. These technologies power Facebook products like News Feed, Instagram, Ads, and Search.

Our research areas include Deep Learning, Massive Sparse Data, Behavior Modeling, Graph Learning, Representation Learning, Reinforcement Learning, Optimization, Distributed/GPU Training, Multi-Modality, Visualization, etc.

Latest Publications

Ranking & Recommendations

Feature Selection for Facebook Feed Ranking System via a Group-Sparsity-Regularized Training Algorithm

In this paper, we propose a novel neural-network-suitable feature selection algorithm, which selects important features from the input layer during training.

Ranking & Recommendations

Training with Low-precision Embedding Tables

In this work, we focus on building a system to train continuous embeddings in low precision floating point representation.

Ranking & Recommendations

Horizon: Facebook's Open Source Applied Reinforcement Learning Platform

In this paper we present Horizon, Facebook’s open source applied reinforcement learning (RL) platform.

Join Us

Tackle the world's most complex technology challenges.

Join Our Team

Latest News

Visit the AI Blog for updates on recent publications, new tools, and more.

Visit Blog