Plus: Google’s Virtual Brain on the Lookout for Cats and a New Robot Dominates Rock-Paper-Scissors.

Neural Network Looks for Cats

A sophisticated neural network constructed by the secretive Google X Lab, which is also responsible for the self-driving car and augmented-reality glasses, has been turned loose on the Internet to test machine learning and advance the concepts of artificial intelligence by doing what millions of humans do each day: search for cat videos.

The system is composed of 16,000 computer processors linked through 1 billion connections, representing one of the largest neural networks in the world. In attempting to have the virtual brain more closely resemble the functioning of a human cortex, the researchers applied an algorithm to see how accurately the system could identify and sort images.

“The ‘brain’ simulation was exposed to 10 million randomly selected YouTube video thumbnails over the course of three days and, after being presented with a list of 20,000 different items, it began to recognize pictures of cats using a ‘deep learning’ algorithm. This was despite being fed no information on distinguishing features that might help identify one,” Wired.com explains. “Picking up on the most commonly occurring images featured on YouTube, the system achieved 81.7 percent accuracy in detecting human faces, 76.7 percent accuracy when identifying human body parts and 74.8 percent accuracy when identifying cats.”

Unlike traditional image-recognition technology, Google’s neural network was not directed to look for specific features of a target object before being presented with any images. As it scanned through millions of pictures, it began to react to specific elements within those pictures, much as the pathways for particular objects or stimuli are built up in the human brain.

“The research is representative of a new generation of computer science that is exploiting the falling cost of computing and the availability of huge clusters of computers in giant data centers. It is leading to significant advances in areas as diverse as machine vision and perception, speech recognition and language translation,” the New York Times reports. “Although some of the computer science ideas that the researchers are using are not new, the sheer scale of the software simulations is leading to learning systems that were not previously possible.”

The Unbeatable Rock-Paper-Scissors Robot

In yet another warning sign of the coming machine revolution, Japanese scientists have developed a robot that plays rock-paper-scissors against human beings and always wins. Always.

Developers from the Ishikawa Oku Laboratory in Tokyo designed the device to be a “human-machine cooperation system,” but it hardly wants to cooperate with us. The rock-paper-scissors robot seeks only to win against its flesh-and-blood opponents, and so far it has proven unbeatable.

“A high-speed camera captures its human opponent’s choice of the three weapons, and it takes only a millisecond for the robot to form the winning hand shape,” CNET News explains. “…the hand looks like it’s almost playing honestly and showing its weapon at the same time as the hapless human. By the time he or she has realized what’s going on, the robot has already won. It wins 100 percent of the time, according to the researchers.”

The robot is able to perceive and react to its competitor at speeds that far outpace human capabilities, giving it somewhat of an unfair advantage. It can see its opponent’s move and respond to it in less than 2 milliseconds; by comparison, the average person takes 100 milliseconds to blink an eye.

All hail our robot masters:

Average Fuel-Burning Car vs. Average Fuel-Burning Human

Automobiles and people aren’t all that different from one another: both rely on fuel that is consumed and converted into energy, and both generate waste in the process. But which is the more efficient energy production system?

Social responsibility resource GOOD has put together a handy infographic comparing the average human with the average car. The average person consumes 2,000 calories per day (roughly 16 percent of a single gallon of gas) and burns 4.6 calories per minute moving at 3 miles per hour.

Meanwhile, a gallon of gas for the average car represents 31,268 calories, and a vehicle burns 1,125 calories per minute moving at 60 mph. Moreover, the average person produces .23 pounds of waste each day, while the average car generates 29.6 pounds of CO₂ a day.

Click image for full-size view. Credit: GOOD

Have a great weekend, folks.