Applied Machine Learning

    Facebook Researchers Focus on the Most Challenging Machine Learning Questions at ICML 2016
    by Jason Weston, Leon Bottou, Joaquin Quinonero Candela, Hussein Mehanna, Pierre Andrews, Aditya Kalro, Alexander Sidorov, Ronan Collobert, Armand Joulin, Laurens van der Maaten, David Grangier, Tomas Mikolov, Antoine Bordes, Rob Fergus, Lars Backstrom, Ross Girshickمنذ حوالي ‏3‏ أشهر
    Facebook AI Research (FAIR)
    Blog
    Facebook researchers are actively engaged at the International Conference on Machine Learning (ICML) 2016 being held in New York City this week. Widely known as the leading Machine Learning...
    Improved Arabic Dialect Classification with Social Media Data
    by Fei Huangمنذ عام تقريبا
    Applied Machine Learning
    Publications
    Arabic dialect classification has been an important and challenging problem for Arabic language processing, especially for social media text analysis and machine translation. In this paper we propose...
    Web-Scale Training for Face Identification
    by Yaniv Taigman, Ming Yang, Marc'Aurelio Ranzato, Lior Wolfمنذ عام تقريبا
    Facebook AI Research (FAIR)
    Publications
    We study face recognition and show that three distinct properties have surprising effects on the transferability of deep convolutional networks (CNN)

    Highlights

    Learning to Refine Object Segments
    by Pedro O. Pinheiro, Tsung-Yi Lin, Ronan Collobert, Piotr DollarOctober 10, 2016
    Publication
    A MultiPath Network for Object Detection
    by Sergey Zagoruyko, Adam Lerer, Tsung-Yi Lin, Pedro O. Pinheiro, Sam Gross, Soumith Chintala, Piotr DollarSeptember 18, 2016
    Publication
    Synergy of Monotonic Rules
    by Vladimir Vapnik, Rauf IzmailovAugust 16, 2016
    Publication
    Facebook Researchers Focus on the Most Challenging Machine Learning Questions at ICML 2016
    by Jason Weston, Leon Bottou, Joaquin Quinonero Candela, Hussein Mehanna, Pierre Andrews, Aditya Kalro, Alexander Sidorov, Ronan Collobert, Armand Joulin, Laurens van der Maaten, David Grangier, Tomas Mikolov, Antoine Bordes, Rob Fergus, Lars Backstrom, Ross GirshickJune 19, 2016
    Blog post

    About Applied Machine Learning

    We use machine learning across a diverse set of applications to help people discover better content more quickly, and to connect with the things that matter most to them. We strive to find ways to deliver more engaging content in News Feed, rank search results more accurately, and present the most relevant ads possible. We share many of our discoveries back with the academic community through conferences and publications.

    Our scale often forces us to approach machine learning challenges from a system engineering standpoint, pushing the boundaries of scalable computing and tying together numerous complex platforms to build models that leverage trillions of actions. Our research, and production implementations, leverage many of the innovations being generated from Facebook's research in Distributed Computing, Artificial Intelligence, and Databases and run on the same hardware and network specifications that are being open sourced through the Open Compute project.

    Publications

    Learning to Refine Object Segments
    Pedro O. Pinheiro, Tsung-Yi Lin, Ronan Collobert, Piotr Dollar
    ECCV
    Oct 10, 2016
    In this work we propose to augment feedforward nets for object segmentation with a novel top-down refinement approach.
    A MultiPath Network for Object Detection
    Sergey Zagoruyko, Adam Lerer, Tsung-Yi Lin, Pedro O. Pinheiro, Sam Gross, Soumith Chintala, Piotr Dollar
    BMVC
    Sep 18, 2016
    We test three modifications to the standard Fast R-CNN object detector to determine if they can overcome the object detection challenges in a COCO...
    Synergy of Monotonic Rules
    Vladimir Vapnik, Rauf Izmailov
    JMLR
    Aug 16, 2016
    This article describes a method for constructing a special rule (we call it synergy rule) that uses as its input information the outputs (scores) of...

    Blog

    Facebook Researchers Focus on the Most Challenging Machine Learning Questions at ICML 2016
    by Jason Weston, Leon Bottou, Joaquin Quinonero Candela, Hussein Mehanna, Pierre Andrews, Aditya Kalro, Alexander Sidorov, Ronan Collobert, Armand Joulin, Laurens van der Maaten, David Grangier, Tomas Mikolov, Antoine Bordes, Rob Fergus, Lars Backstrom, Ross GirshickJun 19, 2016

    Join Us

    Do you want to help more than a billion people all over the world connect and share?

    View Open Positions

    Code

    Learn about our open source tools and technologies, our challenging scaling experiences, and more.

    Go to Facebook Code