MIT researchers have devised a way to train semantic parsers by mimicking the way a child learns language. “The system observes captioned videos and associates the words with recorded actions and objects,” ZDNet reports, citing the paper presented this week. “It could make it easier to train parsers, and it could potentially improve human interactions with robots.” From the report: To train their parser, the researchers combined a semantic parser with a computer vision component trained in object, human and activity recognition in video. Next, they compiled a dataset of about 400 videos depicting people carrying out actions such as picking up an object or walking toward an object. Participants on the crowdsourcing platform Mechanical Turk to wrote 1,200 captions for those videos, 840 of which were set aside for training and tuning. The rest were used for testing. By associating the words with the actions and objects in a video, the parser learns how sentences are structured. With that training, it can accurately predict the meaning of a sentence without a video.

Read more of this story at Slashdot.

Source: Slashdot – Can a Robot Learn a Language the Way a Child Does?

Doctors in Switzerland have used an electrical device to help three paralyzed men walk again. The device was inserted around the men’s spines to boost the signals from their brains to their legs. The study has been published in the journal Nature. The BBC reports: The first patient to be treated was 30-year-old Swiss man David M’zee, who suffered a severe spinal injury seven years ago in a sporting accident. David’s doctor said he would never walk again. However, thanks to an electrical implant developed by a team at Ecole Polytechnique Federale de Lausanne (EPFL), he can walk more than half a mile with the implant turned on. Two other men have also managed to walk again, to varying degrees. Gertan Oskan, a 35-year-old engineer from the Netherlands, was knocked over by a car seven years ago. His doctors told him on his birthday that he would be paralyzed for life. He is now beginning to regain some movement. Sebastian Tobler, a 48-year-old man from Germany, was a keen cyclist who loved being out in the countryside before he was knocked off his bike. Now he’s back on a specially adapted bike that is powered mostly by his hands — but also partly by his legs.

Read more of this story at Slashdot.

Source: Slashdot – Spinal Implant Helps Three Paralyzed Men Walk Again

According to new research, the average cost households pay for cable is now up to $107 a month — that’s a 50% increase since 2010 when cable bills were $71.24 a month. When compared to last year, it’s only a 1% increase, “thanks in large part to increasing fees for things like regional sports licensing and taxes,” reports Streaming Observer. From the report: Leichtman Research Group’s data was gathered through a telephone survey of 1,152 households from throughout the United States. The research found that 78% of American households still subscribe to a paid TV subscription. That percentage is down from 86% in 2013, 87% in 2008, and 81% in 2004, but 78% is still a pretty high figure given how high cable costs continue to rise each year and how affordable streaming video services are in comparison.

Read more of this story at Slashdot.

Source: Slashdot – The Average Cable Bill Has Increased More Than 50 Percent Since 2010

With Alaska’s gubernatorial election coming up, Business Insider brings up a report from earlier this year which finds that the Alaska Permanent Fund Dividend — the only large-scale universal basic income program in the U.S. — doesn’t increase unemployment like many feared. An anonymous reader shares the report: The vast majority of Alaska’s roughly 740,000 citizens support the dividend, which gives virtually every citizen an annual check of about $1,000 to $2,000 (that’s $4,000 to $8,000 for a family of four), and both political parties in the state are in favor. Alaskans’ feelings about this universal cash transfer are supported by the findings of a working paper published in February that was written by University of Chicago Harris School of Public Policy professor Damon Jones and University of Pennsylvania School of Public Policy and Practice professor Ioana Marinescu — the annual dividend does not realize fears that such a program would lead people to quit their jobs, lowering employment.

An additional $8,000 for a family is certainly not going to replace a livable income, but, as Jones and Marinescu noted in their paper, studies around a cash assistance experiment in the 1970s, lottery winnings, and a permanent fund dividend for the Eastern Band of Cherokee Indians reduced earned income, and critics of any universal basic income programs have pointed to such findings as proof that anything on a larger scale would be a disaster. But Jones and Marinescu found instead that the larger scale of the program is what allows it to work, and not dissuade people out of the work force. More specifically, Jones and Marinescu determined that part-time employment increased by 17% only in the non-tradable sector (jobs whose output isn’t traded internationally), and that overall employment wasn’t affected because more spending money results in more demand, and thus more jobs.

Read more of this story at Slashdot.

Source: Slashdot – Alaska’s Universal Basic Income Doesn’t Increase Unemployment