Tuesday, September 02, 2014
Search
  
Submit your own News for
inclusion in our Site.
Click here...
Breaking News
webOS Is Still Alive With LuneOS ROM Release for Android, webOS Devices
AMD Launches AMD Radeon R9 285 Graphics, "Never Settle: Space Edition" Game Bundle
AMD Introduces New 8-core FX-series Processors
New Philips Hue Beyond Combines Functionality And Ambient Lighting for Home
LG and Samsung Add Swarovski Crystals on Their Products
Pioneer DDJ-WeGO3 Allows You To Mix Tracks from Spotify or iTunes
Apple's iCloud Could Have Allowed Celebrity Nude-Photo Leak
Google To Launch New Budget Phone In India
Active Discussions
help questions structure DVDR
Made video, won't play back easily
Questions durability monitor LCD
Questions fungus CD/DVD Media, Some expert engineer in optical media can help me?
CD, DVD and Blu-ray burning for Android in development
IBM supercharges Power servers with graphics chips
Werner Vogels: four cloud computing trends for 2014
Video editing software.
 Home > News > General Computing > Google ...
Last 7 Days News : SU MO TU WE TH FR SA All News

Tuesday, June 26, 2012
Google Labs Create Computer That Mimics Human Brain


Google researchers have built a 9-layer neural network that can learn to detect faces using only unlabeled images.

The researchers built a neural network, which mimics the working of a biological brain, that worked out how to spot pictures of cats in just three days.

The cat-spotting computer was created as part of a larger project to investigate machine learning.

The model has 1 billion connections and the dataset has 10 million 200x200 pixel images randomly downloaded from the internet. It was trained on a cluster of 1,000 machines (16,000 cores) for three days.

The results show it's possible to train a face detector without having to label images as containing a face or not. The detector is robust to translation, scaling, and out-of-plane rotation. It's also sensitive to other high-level concepts such as cat faces and human bodies.

According to Google, the network obtained 15.8% accuracy in recognizing 20,000 object categories, a leap of 70% relative improvement over the previous state-of-the-art.

The team is presenting a paper on its findings at the International Conference on Machine Learning that is being held in Edinburgh from 25 June to 1 July.


Previous
Next
Judge Blocks U.S. Sales of Samsung Galaxy Tab 10.1        All News        Zynga To Offer New Games, Social Network, API For Game Developers
Yahoo Teams Up With Spotify For On-demand Music Service     General Computing News      SanDisk Acquires Storage Software Maker Schooner Information Technology

Get RSS feed Easy Print E-Mail this Message

Related News
Google Tests Drones For Delivery Of Goods
German Regulator Will Pursue Complaint Against Publishers
Google Buys JetPac
Google Releases Fit SDK, Special Version Of Android L
Google To Rank Websites with HTTPS Higher
Google and Barnes & Noble Take On Amazon
Google To Unveil Instagram Competitor
Google's Ambitious Secret Barge Project Was Scarped
Google To Show Ratings to Search-Results Ads
Google Does Not Confirm to $1 Billion Acquisition Of Twitch
Google Implemented "Right To Be Forgotten" Ruling Inadequately, EC says
European Regulators May Need Extra Concessions From Google

Most Popular News
 
Home | News | All News | Reviews | Articles | Guides | Download | Expert Area | Forum | Site Info
Site best viewed at 1024x768+ - CDRINFO.COM 1998-2014 - All rights reserved -
Privacy policy - Contact Us .