Artificial Intelligence
Computers learn language (and world domination) by reading the manual
Researchers at MIT's Computer Science and Artificial Intelligence Lab have been able to create computers that learn language by doing something that many people consider a last resort when tackling an unfamiliar task - reading the manual (or RTBM). Beginning with virtually no prior knowledge, one machine-learning system was able to infer the meanings of words by reviewing instructions posted on Microsoft's website detailing how to install a piece of software on a Windows PC, while another was able to learn how to play Sid Meier's empire-building Civilization II strategy computer game by reading the gameplay manual. Read More
It's probably still going to be a while before autonomous, self-aware androids are wandering amongst us. That scenario has come a little closer to reality, however, with researchers from the University of Southern California having created a functioning synapse circuit using carbon nanotubes. An artificial version of the connections that allow electrical impulses to pass between neurons in our brains, the circuit could someday be one component of a synthetic brain. Read More
Standing around, sorting debris from construction or demolition projects for recycling ... it probably isn't anyone's idea of a good time. Given the risks posed by sharp or heavy objects and airborne particles, it's not a particularly safe way of making a buck either. That's where, perhaps someday soon, ZenRobotics' Recycler autonomous waste-sorting robot arm will come into play. Read More
Researchers have developed an "anytime" universal intelligence test – a test that can be interrupted at any time and continued later, but that gives a more accurate idea of the intelligence of the test subject. The test, developed by researchers working in Spain and Australia, can be applied to any subject, whether biological or not, at any point in its development (child or adult), for any system now or in the future, and with any level of intelligence or speed, making it ideal for evaluating the progress of artificial intelligence systems. Read More
Learning to crawl before they walk makes for smarter robots
Everyone knows that, unless you’re extraordinarily gifted, you need to crawl before you can walk. Turns out the same principle could also apply to robots. In a first-of-its-kind experiment conducted by University of Vermont (UVM) roboticist Josh Bongard created both simulated and physical robots that, like tadpoles becoming frogs, change their body forms while learning how to walk. He found that these evolving robots were able to learn more rapidly than ones with fixed body forms and that, in their final form, the changing robots had developed a more robust gait. Read More
Getting a computer to understand what a person is saying is one thing, but getting it to understand how they’re saying it is another. If we’re ever to see a system that truly comprehends the meaning behind the words – and not just the words themselves – then such systems will need to be able to put the words in context. Researchers from Spain’s Universidad Politécnica de Madrid are trying to achieve this by developing an application that allows computers to accurately recognize human emotions based on automated voice analysis. Read More
Currently, computer search and classification of images is based on the name of the file or folder or on features such as size and date. That’s fine if the name of the file reflects its content but isn’t much good when the file is given an abstract name that only holds meaning to the person providing it. This drawback means companies in the search business, such as Google and Microsoft, are extremely interested in giving computers the ability to automatically interpret the visual contents and video. A technique developed by the University of Granada does just that, allowing pictures to be classified automatically based on whether individuals or specific objects are present in the images. Read More
If they were real, the Transformers harking from Cybertron would be considered pretty remarkable pieces of machinery. But their transforming abilities are limited to just two forms. By combining origami and electrical engineering, researchers at MIT and Harvard are working to develop the ultimate reconfigurable robot – one that can turn into absolutely anything. To test out their theories, the researchers built a prototype that can automatically assume the shape of either an origami boat or a paper airplane when it receives different electrical signals. Read More
A robot guided by 3-D ultrasound and artificial intelligence has demonstrated it can locate lesions in simulated breast and prostate tissue and take biopsies without human assistance. A team of bioengineers at Duke University, North Carolina, 'souped up' an existing robot arm with a purpose-built ultrasound system which acts as the robot's 'eyes' by collecting data from its scan and locating its target. An artificial intelligence program processes the real-time 3D information from the ultrasound and gives the robot specific commands to perform using a mechanical 'hand' that can manipulate the same biopsy plunger device used by doctors. Read More
Most airplane landings are less than graceful. The aircraft slowly maneuvers into an approach pattern, begins a long descent, and then slams on the brakes as soon as it touches down, which barely seems to barely bring it to a rest a mile later. Birds, however, can switch from barreling forward at full speed to lightly touching down on a target as narrow as a telephone wire. MIT researchers have now given a foam glider this same ability using a new control system that could have important implications for robotic planes, greatly improving their maneuverability and potentially allowing them to recharge their batteries simply by alighting on power lines. Read More