Technology Review reports that a team of eight artificial-intelligence specialists at Facebook is working on a technique dubbed “deep learning,” which runs on a system of simulated brain cells used to make more precise judgements about a user’s text, photos, and future behavior.
Mike Schroepfer, Facebook’s chief technology officer, says the goal is to cut down the number of posts served by Facebook’s personalized News Feed from 1,500 or so to just 30 to 60 of the most relevant posts. One reason for making the News Feed smarter and more personalized: On mobile devices, which account for more and more of Facebook’s usage, people visit more frequently but more briefly.
- Facebook Wants To Know What Your Posts Really Mean (readwrite.com)
- Facebook developing ‘deep learning’ AI to understand user posts (electronista.com)
- Facebook Turns to Artificial Intelligence in an Attempt to Improve Your Brain-Dead News Feed (betabeat.com)
- Facebook working on “deep learning” artificial intelligence to improve news feed (slashgear.com)
- Facebook Launches Advanced AI Effort to Find Meaning in Your Posts (technologyreview.com)
- Personalization and the Facebook News Feed (redbricksmedia.com)
- London Gets To House Facebook’s First Non-US Engineering Centre (misco.co.uk)
- Facebook Chases Google’s Deep Learning with New Research Group | MIT Technology Review (ghanaictnews.wordpress.com)
- Facebook launches advanced AI effort to find meaning in your posts (salon.com)
- Facebook developing brain-like AI to find deeper meaning in feeds and photos (engadget.com)