Papers of the day   All papers

TOWARDS FEDERATED LEARNING AT SCALE: SYSTEM DESIGN

Comments

Feb 05 2019 Andrew Trask

I'm *extremely* excited to share one of the most *exciting* papers of the year. Yesterday - Google published how they do #FederatedLearning at scale on tens of millions of mobile phones #privacy is real - and Google is a leader! https://arxiv.org/pdf/1902.01046.pdf #100DaysOfMLCode #ppml https://t.co/CXz8JfAIBN
17 replies, 1651 likes


Aug 02 2019 Andrew Trask

.@Google recently published how they do #FederatedLearning at scale on tens of millions of mobile phones #privacy is happening!!! https://arxiv.org/pdf/1902.01046.pdf #100DaysOfMLCode #100DaysOfCode #ppml https://t.co/nLwwO3NZuF
7 replies, 618 likes


May 08 2019 DataScienceNigeria

Great views on Federated Learning at #IO19, a leap forward for on-device machine learning. Now, we can train AI models without data ever leaving our devices & Gboard can learn new words Imagine possibilities for Health &Agriculture in Nigeria https://arxiv.org/pdf/1902.01046.pdf https://t.co/TSJLPuhzao
2 replies, 40 likes


Aug 02 2019 Daniel Situnayake

If it works for your application, on-device ML is just plain better. Superior user privacy, no need to handle sensitive data, less backend engineering
0 replies, 32 likes


Aug 02 2019 William Falcon

lol. pitched this to an advisor @Columbia when I was an undergrad. they laughed me out of the room haha. almost joined @clarifai to help build something similar, but instead launched my own startup which worked out well 🙃... #makeYourOwnLuck
1 replies, 28 likes


May 13 2019 Mariya Yao

Google shows how it brings "the code to the data, instead of the data to the code", producing a scalable production system for Federated Learning. Check out the paper if you want to dive deeper into the design they introduce. https://arxiv.org/pdf/1902.01046.pdf?utm_campaign=meetedgar&utm_medium=social&utm_source=meetedgar.com #AI #ML
0 replies, 12 likes


May 13 2019 TOPBOTS

Google shows how it brings "the code to the data, instead of the data to the code", producing a scalable production system for Federated Learning. Check out the paper if you want to dive deeper into the design they introduce. https://arxiv.org/pdf/1902.01046.pdf?utm_campaign=meetedgar&utm_medium=social&utm_source=meetedgar.com #AI #ML
0 replies, 3 likes


Feb 13 2019 Iryo Network

Google is working on machine learning with distributed data, where analyses are done locally, thus guaranteeing anonymity. It's a version of the "analyse in place" approach we have been defending for some time now. https://arxiv.org/pdf/1902.01046.pdf
0 replies, 2 likes


Aug 02 2019 Arjun S

@GoogleAI recently published an amazing paper on building federated learning systems at scale! https://arxiv.org/pdf/1902.01046.pdf #federatedlearning #machinelearning #differentialprivacy
0 replies, 2 likes


Aug 02 2019 Petteri Teikari

Nice :)
0 replies, 1 likes


Content