RESEARCH

Machine Learning at Facebook: Understanding Inference at the Edge

February 16, 2019

Abstract

At Facebook, machine learning provides a wide range of capabilities that drive many aspects of user experience including ranking posts, content understanding, object detection and tracking for augmented and virtual reality, speech and text translations. While machine learning models are currently trained on customized datacenter infrastructure, Facebook is working to bring machine learning inference to the edge. By doing so, user experience is improved with reduced latency (inference time) and becomes less dependent on network connectivity. Furthermore, this also enables many more applications of deep learning with important features only made available at the edge. This paper takes a data-driven approach to present the opportunities and design challenges faced by Facebook in order to enable machine learning inference locally on smartphones and other edge platforms.

Download the Paper

Related Publications

May 17, 2019

COMPUTER VISION

SPEECH & AUDIO

GLoMo: Unsupervised Learning of Transferable Relational Graphs | Facebook AI Research

Modern deep transfer learning approaches have mainly focused on learning generic feature vectors from one task that are transferable to other tasks, such as word embeddings in language and pretrained convolutional features in vision. However,…

Zhilin Yang, Jake (Junbo) Zhao, Bhuwan Dhingra, Kaiming He, William W. Cohen, Ruslan Salakhutdinov, Yann LeCun

May 17, 2019

May 10, 2019

Insights on Visual Representations for Embodied Navigation Tasks | Facebook AI Research

Recent advances in deep reinforcement learning require a large amount of data and result in representations that are often over specialized to the target task. In this work, we study the underlying potential causes for this specialization by…

Erik Wijmans, Julian Straub, Dhruv Batra, Judy Hoffman, Sayna Ebrahimi

May 10, 2019

May 06, 2019

COMPUTER VISION

NLP

No Training Required: Exploring Random Encoders for Sentence Classification | Facebook AI Research

We explore various methods for computing sentence representations from pre-trained word embeddings without any training, i.e., using nothing but random parameterizations. Our aim is to put sentence embeddings on more solid footing by 1) looking…

John Wieting, Douwe Kiela

May 06, 2019

May 13, 2019

Soft Labeling in Stochastic Shortest Path Problems | Facebook AI Research

The Stochastic Shortest Path (SSP) problem is an established model for goal-directed probabilistic planning. Despite its broad applicability, wide adoption of the model has been impaired by its high computational complexity. Efforts to address…

Luis Pineda, Shlomo Zilberstein

May 13, 2019

Help Us Pioneer The Future of AI

We share our open source frameworks, tools, libraries, and models for everything from research exploration to large-scale production deployment.