AI + dyslexia. Facebook using ML to source local news. Future of voice. http://cognitionx.com/news-briefing/
We’re taking a pause on CES updates today, but will be back tomorrow with more highlights.
Today, we wanted to focus on the impact that AI can (and is) having on the environment. From predicting thunderstorms to protecting threatened species of animals in their natural habitats, the potential is massive. Although these stories don’t make it to the front page of the newspaper, they are tremendously important.
Now, a company called Lexplore has created an artificially intelligent helper to make it easier than ever for schools to spot students struggling with the learning disability. It does this by tracking the movement of kids eyes as they read.
Lexplore was spun-out of the Karolinska Institute in Stockholm (the same institute that hands out the Nobel Prize in Medicine). It’s built on research by Gustaf Öqvist Seimyr and Mattias Nilsson Benfatto who applied cutting-edge tracking and machine learning technologies to discoveries from earlier studies on eye movement.
Facebook wants to make it easier for people to find local news from vetted sources. The social network is testing a new section inside its app called “Today In,” a feed made up entirely of local news, events and announcements.
Facebook is using machine-learning software to surface content in this new section. Local news publishers who appear there will all be approved and vetted by the company’s News Partnerships team, which is overseen by former NBC news anchor Campbell Brown, according to a company spokesperson.
Federal regulators are taking the first step toward creating a policy guiding the development of autonomous transportation beyond self-driving cars to include trucks, buses, and other ground-based modes. The US Transportation Department will soon publish four requests for public comment on how to cast aside roadblocks for transportation advancements in vehicles, trains, buses, commercial trucking and transit systems, Transportation Secretary Elaine Chao said Wednesday.
“Right now there are too many outdated transportation rules, terms and concepts that no longer apply to an automated world,” Chao said during a speech at CES, the annual technology show sponsored by the Consumer Technology Association in Las Vegas. “This request for input will help the government identify which regulations, parts of regulations or terminology need to be updated to allow for innovation to move forward.”
A patent application published January 4 details how Google could use “optical sensors” placed in patients’ devices or belongings to capture data on individual’s cardiovascular function – all with the aim of motivating behavioral changes and reducing instances of heart disease.
The sensors might even be positioned (per the patent’s illustrations) in a “sensing milieu” in a patient’s bathroom. Google’s new invention for at-home health tracking would
monitor certain aspects of a patient’s physical appearance; and
track changes in appearance that relate to cardiovascular health problems.
Scientists have already used CRISPR gene editing techniques try and treat or cure diseases, improve birth rates and even alleviate allergic reactions. The problem is that it can be tricky to edit DNA accurately and safely. A new AI project from Microsoft called Elevation uses a ton of CRISPR data and machine learning to predict where best to edit a strand of DNA to alleviate side effects and speed up the editing process itself.
Many of the regions in genetic material look the same, which means that CRISPR can often work on the wrong gene, causing what scientists call off-target effects. A new paper published by researchers from Microsoft, MIT, Harvard, UCLS and Massachusetts General Hospital describes the new tool that predicts these effects.
Researchers at Carnegie Mellon University and the University of North Carolina at Chapel Hill have developed a method of creating inconspicuous eyeglasses that can be used to thwart identification by facial recognition algorithms, according to their published findings.
The researchers developed five pairs of “universal” glasses that “facilitate misclassification” through an adversarial generative nets (AGN) method, which involves using neural networks to produce designs with different colors and textures for glasses that can either evade correct identification or impersonate a specific target. They say the glasses can be manufactured with a 3D printer, and can be used by roughly 90 percent of the population to fool deep neural network-based facial recognition. The researchers consider the method to be inconspicuous, scalable, and robust against some defenses.