< < Anut21ng via Getty ImagesIt’s easy to be worried when you cough these days — is it COVID-19, or are you just clearing your throat? You might get a clearer answer soon. MIT researchers have developed AI that can recognize forced coughing from people who have COVID-19, even if they’re otherwise asymptomatic. The trick was to develop a slew of neural networks that can distinguish subtle changes indicative of the novel coronavirus’ effects.One neural network detects sounds associated with vocal strength. Another listens for emotional states that reflect a neurological decline, such as increased frustration or a “flat affect.” A third network, meanwhile, gauges changes in respiratory performance. Throw in an algorithm that checks for muscular degradation (that is, weaker coughs) and it provides a more complete picture of someone’s health.The AI is highly accurate in early tests. After the team trained its model on tens of thousands of cough and dialog samples, the technology recognized 98.5 percent of coughs from people with confirmed COVID-19 cases. It identified 100 percent of people who were ostensibly asymptomatic, too.There are clear limits. The technology isn’t meant to diagnose symptomatic people, as they might have other conditions that produce similar behavior. And while it’s quite capable, you wouldn’t want to use this for a definitive verdict on whether or not you’re infected.This isn’t a theoretical exercise, though. The scientists are developing a “user-friendly” app that could be used as a prescreening tool for the virus. You might only have to cough into your phone each day to determine if it’s safe for you to head outside. The researchers even suggest this could put an end to pandemics if the tool was always listening in the background, although that’s a big “if” when it would likely raise privacy issues.

<

 

Anut21ng via Getty Images

It’s easy to be worried when you cough these days — is it COVID-19, or are you just clearing your throat? You might get a clearer answer soon. MIT researchers have developed AI that can recognize forced coughing from people who have COVID-19, even if they’re otherwise asymptomatic. The trick was to develop a slew of neural networks that can distinguish subtle changes indicative of the novel coronavirus’ effects.

One neural network detects sounds associated with vocal strength. Another listens for emotional states that reflect a neurological decline, such as increased frustration or a “flat affect.” A third network, meanwhile, gauges changes in respiratory performance. Throw in an algorithm that checks for muscular degradation (that is, weaker coughs) and it provides a more complete picture of someone’s health.

The AI is highly accurate in early tests. After the team trained its model on tens of thousands of cough and dialog samples, the technology recognized 98.5 percent of coughs from people with confirmed COVID-19 cases. It identified 100 percent of people who were ostensibly asymptomatic, too.

There are clear limits. The technology isn’t meant to diagnose symptomatic people, as they might have other conditions that produce similar behavior. And while it’s quite capable, you wouldn’t want to use this for a definitive verdict on whether or not you’re infected.

This isn’t a theoretical exercise, though. The scientists are developing a “user-friendly” app that could be used as a prescreening tool for the virus. You might only have to cough into your phone each day to determine if it’s safe for you to head outside. The researchers even suggest this could put an end to pandemics if the tool was always listening in the background, although that’s a big “if” when it would likely raise privacy issues.

< < TikTok’s attempt to sell itself and avert a possible US ban may run into some complications. The Wall Street Journal reports that China has unveiled new restrictions on AI technology exports that could affect TikTok. The new rules bar the exports of tech like content suggestions, text analysis and voice recognition unless a company receives a license — technology TikTok uses in some cases.The Chinese government has issued a not-so-subtle warning to TikTok parent ByteDance in turn. Government advisor Cui Fan told the state-run Xinhua News Agency that ByteDance should “seriously and cautiously” consider stopping its sales talks for TikTok. Even if ByteDance no longer has a stake in TikTok, there would probably be some technology transfers that could violate the rules, the advisor said.The country’s Ministry of Commerce argued that the export list changes were overdue after remaining the same since 2008. It was important given the breakneck pace of technological improvement and China’s increasingly competitive output, according to officials.Neither ByteDance nor the Commerce Ministry has commented on the new rules.The move escalates an already intense dispute between China and the US. The two sides are already locked in a trade war, and the US has already implemented trade restrictions on companies like Huawei and ZTE over alleged security risks. The pressure on TikTok to drop ByteDance is an extension of this. In that regard, it’s not surprising that China is countering with tighter export limits. This theoretically pressures the US to make concessions and allow more access to Chinese tech.

<

 

TikTok’s attempt to sell itself and avert a possible US ban may run into some complications. The Wall Street Journal reports that China has unveiled new restrictions on AI technology exports that could affect TikTok. The new rules bar the exports of tech like content suggestions, text analysis and voice recognition unless a company receives a license — technology TikTok uses in some cases.

The Chinese government has issued a not-so-subtle warning to TikTok parent ByteDance in turn. Government advisor Cui Fan told the state-run Xinhua News Agency that ByteDance should “seriously and cautiously” consider stopping its sales talks for TikTok. Even if ByteDance no longer has a stake in TikTok, there would probably be some technology transfers that could violate the rules, the advisor said.

The country’s Ministry of Commerce argued that the export list changes were overdue after remaining the same since 2008. It was important given the breakneck pace of technological improvement and China’s increasingly competitive output, according to officials.

Neither ByteDance nor the Commerce Ministry has commented on the new rules.

The move escalates an already intense dispute between China and the US. The two sides are already locked in a trade war, and the US has already implemented trade restrictions on companies like Huawei and ZTE over alleged security risks. The pressure on TikTok to drop ByteDance is an extension of this. In that regard, it’s not surprising that China is countering with tighter export limits. This theoretically pressures the US to make concessions and allow more access to Chinese tech.

< < Due to the coronavirus pandemic, DARPA will no longer hold an in-person event for its third and final AlphaDogfight Trial that’s scheduled to take place from August 18th to the 20th. It’ll be held virtually instead, and both participants and viewers alike can watch online as artificial intelligence (AI) algorithms control simulated F-16 fighter planes in aerial combat. By the end of the three-day event, viewers will even get to witness a matchup between the top AI and an experienced Air Force fighter pilot who’ll also be controlling a virtual F-16.The Department of Defense agency started the AlphaDogfight Trials to expand its base of AI developers under the Air Combat Evolution (ACE) program. ACE’s purpose is to automate air-to-air combat and to build human pilots’ trust in artificial intelligence, so they can fight side-by-side in the future. The first trial, which was held in November 2019, featured algorithms in early development.DARPA held the second trial in January this year and pitted participants’ greatly improved algorithms against AI adversaries developed by the Johns Hopkins University Applied Physics Laboratory. Both trials were hosted in person at the APL.The third trial will be streamed from the APL, as well, with eight teams flying against the lab’s AI adversary algorithms on the first day and then against each other in a round robin tournament on the second. It’s the first time the participants are pitting their AIs against one another in public. On the last day, the top four AIs will compete in a single-elimination tournament for the championship title. The last team standing will then get the chance to fly against an Air Force fighter pilot controlling a virtual plane.Col. Dan “Animal” Javorsek, program manager in DARPA’s Strategic Technology Office, said in a statement:“Regardless of whether the human or machine wins the final dogfight, the AlphaDogfight Trials is all about increasing trust in AI. If the champion AI earns the respect of an F-16 pilot, we’ll have come one step closer to achieving effective human-machine teaming in air combat, which is the goal of the ACE program.”Teams from Aurora Flight Sciences, EpiSys Science, Georgia Tech Research Institute, Heron Systems, Lockheed Martin, Perspecta Labs, PhysicsAI and SoarTech make up the eight participants for trial 3. Interested viewers will have to register beforehand to able to tune in: US citizens have until August 17th to sign up, while everyone else has until August 11th.

<

 

Due to the coronavirus pandemic, DARPA will no longer hold an in-person event for its third and final AlphaDogfight Trial that’s scheduled to take place from August 18th to the 20th. It’ll be held virtually instead, and both participants and viewers alike can watch online as artificial intelligence (AI) algorithms control simulated F-16 fighter planes in aerial combat. By the end of the three-day event, viewers will even get to witness a matchup between the top AI and an experienced Air Force fighter pilot who’ll also be controlling a virtual F-16.

The Department of Defense agency started the AlphaDogfight Trials to expand its base of AI developers under the Air Combat Evolution (ACE) program. ACE’s purpose is to automate air-to-air combat and to build human pilots’ trust in artificial intelligence, so they can fight side-by-side in the future. The first trial, which was held in November 2019, featured algorithms in early development.

DARPA held the second trial in January this year and pitted participants’ greatly improved algorithms against AI adversaries developed by the Johns Hopkins University Applied Physics Laboratory. Both trials were hosted in person at the APL.

The third trial will be streamed from the APL, as well, with eight teams flying against the lab’s AI adversary algorithms on the first day and then against each other in a round robin tournament on the second. It’s the first time the participants are pitting their AIs against one another in public. On the last day, the top four AIs will compete in a single-elimination tournament for the championship title. The last team standing will then get the chance to fly against an Air Force fighter pilot controlling a virtual plane.

Col. Dan “Animal” Javorsek, program manager in DARPA’s Strategic Technology Office, said in a statement:

“Regardless of whether the human or machine wins the final dogfight, the AlphaDogfight Trials is all about increasing trust in AI. If the champion AI earns the respect of an F-16 pilot, we’ll have come one step closer to achieving effective human-machine teaming in air combat, which is the goal of the ACE program.”

Teams from Aurora Flight Sciences, EpiSys Science, Georgia Tech Research Institute, Heron Systems, Lockheed Martin, Perspecta Labs, PhysicsAI and SoarTech make up the eight participants for trial 3. Interested viewers will have to register beforehand to able to tune in: US citizens have until August 17th to sign up, while everyone else has until August 11th. 

< <New feature can deep-link to specific text on a Web page, with highlighting.Google has been cooking up an extension to the URL general referred to as "textual content Fragments." the brand new link style will allow you to link not just to a web page but to distinct textual content on a page, for you to get scrolled to and highlighted automatically once the web page masses. It's like an anchor hyperlink, however with highlighting and creatable by using anyone.The feature has definitely been supported in Chrome for the reason that version eighty, which hit the stable channel in February. Now a new extension from Google makes it effortless to create this new hyperlink style, in order to work for someone else utilising Chrome on laptop OSes and Android. Google has proposed the thought to the W3C and hopes other browsers will adopt it, however although they don't, the hyperlinks are backward-suitable.The syntax for this URL is pretty unusual watching. After the URL, the magic is within the string "#:~:text=" after which whatever text you need to check. So a full link would seem like this:https://en.Wikipedia.Org/wiki/Cat#:~:textual content=Most breeds of cat have a famous fondness for sitting in high areasfor those who reproduction and paste this into Chrome, the browser will open Wikipedia's cat web page, scroll to the primary textual content that matches "Most breeds of cat have a noted fondness for sitting in high areas," and can spotlight it. If the text would not healthy something, the page will still load. Backward-compatibility works on the grounds that browsers currently help the number sign (#) as a URI fragment, which traditionally gets used for anchor hyperlinks that are made via the web page creator. If you paste this into a browser that does not aid it, the web page will still load, and everything after the number sign will simply be ignored as a foul anchor link. Up to now, so excellent.One predicament is that this means that you could have areas in a URL. On a webpage or discussion board, which you could hand-code the link with a href tag (or something the non-HTML an identical is) and everything will work. For instant messengers and social media although, which don't allow code and use automatic URL parsers, things get a little more tricky. Each URL parser treats a space as the end of a URL, so you'll ought to use percent-encoding to exchange all the areas with the an identical "%20." URL parsers now have a shot at linkifying this accurately, nevertheless it looks like a large number:https://en.Wikipedia.Org/wiki/Cat#:~:textual content=Mostpercent20breedsp.C20of%20cat%20havepercent20a%20notedp.C20fondnesspercent20forpercent20sittingp.C20inpercent20highpercent20places.Areas aren't the one characters that may cause issues. The commonplace RFC 3986 defines a number of "reserved" characters as having a specific which means in a URL, so they is just not in a URL. Internet-page-authoring tools are likely to handle these characters automatically, but now that you are embedding arbitrary sentences in a URL for highlighting, there may be a greater hazard you can run into such a reserved characters:! * ' ( ) ; : @ & = + $ , / ? # [ ]. All of them must be percentage-encoded to ensure that the URL to work, and Google's extension takes care of that for you.Google's new Chrome extension, called "link to textual content Fragment," (it's also on Github) will put a brand new entry in Chrome's correct-click on menu. You simply highlight textual content on a web page, right-click it, and hit "reproduction link to selected textual content." Like magic, a textual content fragment hyperlink will end up for your clipboard. All the textual content encoding is done mechanically, so the hyperlink should work with most web sites and messengers.Google looks as if it will pushing out support for text fragments throughout its internet ecosystem, even without the W3C. The links have already began to exhibit up in some Google search results, which enable Chrome users to zip correct to the relevant textual content. It's generally best a topic of time earlier than hyperlink production moves from an extension to a natural Chrome feature.Checklist picture by using Chrome

<

New feature can deep-link to specific text on a Web page, with highlighting.

Google has been cooking up an extension to the URL general referred to as "textual content Fragments." the brand new link style will allow you to link not just to a web page but to distinct textual content on a page, for you to get scrolled to and highlighted automatically once the web page masses. It's like an anchor hyperlink, however with highlighting and creatable by using anyone.

The feature has definitely been supported in Chrome for the reason that version eighty, which hit the stable channel in February. Now a new extension from Google makes it effortless to create this new hyperlink style, in order to work for someone else utilising Chrome on laptop OSes and Android. Google has proposed the thought to the W3C and hopes other browsers will adopt it, however although they don't, the hyperlinks are backward-suitable.

The syntax for this URL is pretty unusual watching. After the URL, the magic is within the string "#:~:text=" after which whatever text you need to check. So a full link would seem like this:

https://en.Wikipedia.Org/wiki/Cat#:~:textual content=Most breeds of cat have a famous fondness for sitting in high areas

for those who reproduction and paste this into Chrome, the browser will open Wikipedia's cat web page, scroll to the primary textual content that matches "Most breeds of cat have a noted fondness for sitting in high areas," and can spotlight it. If the text would not healthy something, the page will still load. Backward-compatibility works on the grounds that browsers currently help the number sign (#) as a URI fragment, which traditionally gets used for anchor hyperlinks that are made via the web page creator. If you paste this into a browser that does not aid it, the web page will still load, and everything after the number sign will simply be ignored as a foul anchor link. Up to now, so excellent.

One predicament is that this means that you could have areas in a URL. On a webpage or discussion board, which you could hand-code the link with a href tag (or something the non-HTML an identical is) and everything will work. For instant messengers and social media although, which don't allow code and use automatic URL parsers, things get a little more tricky. Each URL parser treats a space as the end of a URL, so you'll ought to use percent-encoding to exchange all the areas with the an identical "%20." URL parsers now have a shot at linkifying this accurately, nevertheless it looks like a large number:

https://en.Wikipedia.Org/wiki/Cat#:~:textual content=Mostpercent20breedsp.C20of%20cat%20havepercent20a%20notedp.C20fondnesspercent20forpercent20sittingp.C20inpercent20highpercent20places.

Areas aren't the one characters that may cause issues. The commonplace RFC 3986 defines a number of "reserved" characters as having a specific which means in a URL, so they is just not in a URL. Internet-page-authoring tools are likely to handle these characters automatically, but now that you are embedding arbitrary sentences in a URL for highlighting, there may be a greater hazard you can run into such a reserved characters:! * ' ( ) ; : @ & = + $ , / ? # [ ]. All of them must be percentage-encoded to ensure that the URL to work, and Google's extension takes care of that for you.

Google's new Chrome extension, called "link to textual content Fragment," (it's also on Github) will put a brand new entry in Chrome's correct-click on menu. You simply highlight textual content on a web page, right-click it, and hit "reproduction link to selected textual content." Like magic, a textual content fragment hyperlink will end up for your clipboard. All the textual content encoding is done mechanically, so the hyperlink should work with most web sites and messengers.

Google looks as if it will  pushing out support for text fragments throughout its internet ecosystem, even without the W3C. The links have already began to exhibit up in some Google search results, which enable Chrome users to zip correct to the relevant textual content. It's generally best a topic of time earlier than hyperlink production moves from an extension to a natural Chrome feature.

Checklist picture by using Chrome