All posts by Daniel Davies

Researchers create computer that learns at the same pace as humans

A computer model has been created that can replicate humans’ unique ability to learn new concepts from a single example.

Researchers at MIT, New York University have managed to develop machine-learning algorithms that can compete with a human’s capacity for learning. The model created by the team is capable of learning and making generalisations about handwritten characters from alphabets,

Lead researcher Brendan Lake acknowledges that that he and the other researchers had a number of reasons for completing their study. “One goal was to better understand human learning,” says Lake. “The other goal was to develop new more human like learning algorithms.

Image courtesy of Brenden Lake

Image courtesy of Brenden Lake

The research team used alphabets from around the world because they sought to directly compare the model they created with human comprehension.

Firstly the model was built to learn a large class of visual symbols, and make generalizations about them from very few examples. This modeling scheme was called the Bayesian Program Learning framework (BPL).

“If we had chosen full visual scenes, such as a street scene in New York City, people would obviously have an advantage over machines because so much of the brain is devoted to visual processing,” says Lake. “If we’d chosen concepts based on financial data then machines would probably have an advantage because they’re better at crunching numbers and these concepts are less intuitive.

“Characters provided a relatively level-playing field. We hoped [that by using alphabets] to generate insight that could generalise to other domains.”

shutterstock_6488956

The researchers hope that the approach taken by the computer model can be broadened so that it can be used for other symbol-based systems, like gestures, dance moves, and the words of spoken and signed languages.

“As we were developing the model, we identified three core ingredients that are important for the model’s performance. The first ingredient is compositionality which is an old idea that representation should be built up from simpler primitives, so in the case of characters these were the pen strokes,” says Lake.

“Another critical ingredient was causality which is representing in an abstract way the causal structure behind where objects come from, and the last key ingredient is learning to learn which is the idea that knowledge or previous concepts can be used to learn new concepts.

“These principles may help to explain how we learn and use other concepts in different domains.”

Read the full paper at sciencemag.org.

FDA approve expanding sponges that can plug life-threatening wounds

The US Food and Drug Administration (FDA) has authorised the use of expanding multi-sponge dressing that can stem hemorrhaging from bullet and non-compressible wounds.

XSTAT 30 wound dressing is an expandable, multi-sponge dressing used to control severe, life-threatening bleeding from wounds in areas that a traditional compression device cannot be placed –such as the groin or armpit. Originally the sponge was designed for the military, but its usage has now been extended to include civilian trauma cases.

“It is exciting to see this technology transition to help civilian first responders control severe, life-threatening bleeding while on the trauma scene,” said acting director of the Office of Device Evaluation in the FDA’s Center for Devices and Radiological Health, William Maisel MD.

trauma-3

According to the United States Army Institute of Surgical Research, 30 to 40% of civilian deaths by traumatic injury are the result of hemorrhaging, and 33 to 56% occur before the patient reaches a hospital.

The sponges work by swelling to fill the wound cavity, and creating a temporary physical barrier to the blood flow.  They remain effective for up to four hours, and allow first response teams to control severe blood loss before the patient is able to receive surgical care. Each applicator can absorb about a pint of blood.

“When a product is developed for use in the battlefield, it is generally intended to work in a worst-case scenario where advanced care might not be immediately available,” said Maisel.

Image courtesy of ffsetla. Feature image courtesy of FDA

Image courtesy of ffsetla. Feature image courtesy of FDA

XSTAT 30 has not been altered in order for it to be used in civilian settings.

On being approved for use with the military, director of the Office of Device Evaluation at the FDA’s Center for Devices and Radiological Health, Christy Foreman said: “XSTAT is a novel device that can be rapidly deployed, providing fast-acting hemorrhage control to stabilize a wounded patient for transport,

“This will be an important new treatment option for our nation’s military to treat injured soldiers who may not be in close proximity to a medical facility.”