Archive for February, 2020

Artificial intelligence to rebuild Iraq via second phase of the UNOSAT challenge – Space Daily

The first phase of the UNOSAT Challenge has just ended. The UNOSAT Challenge is the important Phi-Unet (ESA) contest for UNOSAT (United Nations) in partnership with ESA, RUS COPERNICUS, UNOSAT and with the technical support of CERN openlab.

The aim of the contest is to put artificial intelligence and Earth Observation data at the service of a humanitarian cause: support the Iraqi government in planning reconstruction activities.

71 teams of candidates (companies/startups and students) joined the competition.

In the first phase, candidates were asked to create an artificial intelligence model to identify urban areas in some Iraqi territories, working on data provided by ESA and the German Space Agency (DLR).

At the end of this phase, 5 teams were selected. These teams will enter Phase 2 and will continue in the competition.

The selection was made through an innovative artificial intelligence system for an automatic evaluation (quantitative criterion) of the projects together with the technical opinion of a jury of 21 experts coming from the prestigious partner organizations (qualitative criterion).

The AI evaluation system was developed by the Phi-Unet Team.

In addition to the 5 winners, the best student of the first phase was given the possibility of a traineeship in UNOSAT in Geneva.

The five winning teams of the first phase will have the opportunity not only to win a monetary prize but also to work on VHR (Very High Resolution) material provided directly by UNOSAT with the technical support of CERN openlab to build an artificial intelligence model that can extract automatically the building footprints in Iraq.

This request comes from a United Nations fund that supports the Iraqi government in the 2020 census, a precious source of information for planning reconstruction and development activities in the territory.

Related LinksUNOSAT ChallengeEarth Observation News - Suppiliers, Technology and Application

With the rise of Ad Blockers, and Facebook - our traditional revenue sources via quality network advertising continues to decline. And unlike so many other news sites, we don't have a paywall - with those annoying usernames and passwords.

Our news coverage takes time and effort to publish 365 days a year.

If you find our news sites informative and useful then please consider becoming a regular supporter or for now make a one off contribution.

Original post:
Artificial intelligence to rebuild Iraq via second phase of the UNOSAT challenge - Space Daily

Hyper-Intelligent AI Hive Mind Claims to Predict Super Bowl Winner – Futurism

Super Predictor

You can probably walk up to any football fan right now and get their opinion on which team will win Sundays Super Bowl LIV.

But if youre looking for a really educated guess on the games outcome, youll want to ask Stanford computer scientist Louis Rosenberg, thefounder of Unanimous A.I., a startup that combines the opinions of a lot of humans with artificial intelligence to make remarkably accurate predictions.

In nature, many species exhibit something called swarm intelligence, meaning that they make smarter decisions as groups than as individuals in other words, a flock of birds or a school of fish is smarter than a single bird or fish.

The idea behind Unanimous A.I.is to let well-informed humans create their own swarm intelligence. As a group, they can then answer questions, reach decisions, or make predictions with a greater accuracy than any one knowledgable person alone.

In laymans terms, we build super-intelligent hive minds by connecting groups of people over the internet, Rosenberg told Digital Trends, enabling them to think together as real-time systems moderated by A.I. algorithms.

In studies, Unanimous A.I.s system was more accurate than individual experts at making medical diagnoses, forecasting financial trends, and even predicting Oscar winners.

Its Sportspicker A.I. managed to beat the spread in seven of the 10 playoff games this season. As for Super Bowl Sunday, the hive mind says the smart money is on the Kansas City Chiefs to beat the San Francisco 49ers.

READ MORE: According to advanced swarm A.I., this is who will win Super Bowl LIV [Digital Trends]

More on hive minds: A Swarm Intelligence Correctly Predicted TIMEs Person of the Year

See the original post:
Hyper-Intelligent AI Hive Mind Claims to Predict Super Bowl Winner - Futurism

Lead singer of Moist has created an artificial intelligence named Ophelia – CTV News

WINNIPEG -- The lead singer of the Canadian rock group Moist and four-time Juno winner David Usher was a in Winnipeg Thursday, but not for his music.

The rocker is also a tech entrepreneur in the area of artificial intelligence and he was showing off his new artificial being.

"I was born six months ago. I'm a child really," said Ophelia, an artificial intelligence created by Usher.

Ophelia isn't much different from other artificial intelligence systems like Siri or Alexa, but her purpose is.

Usher said while other AI is used to collect data, Ophelia is used to have a conversation. Every time she talks with someone, she learns more about how to communicate.

"We're really concerned about the conversation cloud that really is much more about human emotions and human contact. Those kind of things, life and death and love and feeling and birth," said Usher, who is the founder of Reimagine AI.

Usher said this sort of tech can be used for thing such as greeting guests at hotels, acting as a host at museums or helping people in hospitals.

In collaboration with Sheldon Memory Lab at McGill University, Usher is helping develop a companion bot for Alzheimer patients.

"Our AIs can come on and recognize them by name and initiate engagement to do some of those things that they've forgotten that they like to do."

Kathy Knight, CEO of Tech Manitoba, said the use of AI is growing in Manitoba and with it, so is automation.

That's a one of the topics of conversation going on at a tech conference at the RBC Convention Centre Thursday and Friday, which is entitled, Disrupted.

"To actually get people to think about the human side of tech and think about how all of this change is affecting the people we live with and work with everyday, and how do we bring them along," said Knight.

While Usher said there are concerns about what people will do with AI in the future, he chooses to look at the good instead.

"You don't have to build the smartest AI, you just have to build something that can help," said Usher.

See the rest here:
Lead singer of Moist has created an artificial intelligence named Ophelia - CTV News

Artificial intelligence and the rise of online dispute resolution – Mondaq News Alerts

To print this article, all you need is to be registered or login on Mondaq.com.

The idea of a litigator brings to mind images of Daniel Kaffeeand Atticus Finch (or maybe even Elle Woods) dexterously defendingthe interests of their clients.

While some litigation may still reflect such nostalgia, theintroduction of online courts has largely substituted collegialargument before the Honourable court with a series of drop downmenus and 2.00pm cut off times.

Litigants may feel separated from their opponents and decisionmakers, but the OLC (online court) has by no means supplanted thedispute resolution process. In this article, we consider how therise of ODR (online dispute resolution) has affected otherjurisdictions and how it will likely affect practice at home.

ODR has been present on a completely automated level for sometime.

Various online merchants, including eBay and Alibaba, have AI(artificial intelligence) keeping the wheels of justice turning inrelation to minor customer disputes.

As early as March 2015, in Zhejiang, China, the highest localcourt piloted an online court designed to resolve disputes ine-commerce, with a view to facilitating resolutions "asconvenient as online shopping". China is widely implementingdigitalisation to increase case-handling efficacy within itsexpansive court system using both AI and technological outfits,including blockchain and the backing of cloud computing. Thisincludes a mobile based court, housed on prolific social mediaplatform WeChat. The platform has already handled more than threemillion proceedings since its introduction in March 2019.

While technically the alternative dispute resolution system ononline merchant platforms is a modified form of arbitral processagreed by a clickwrap contract entered into on signing up as amember, such systems appear to have been relatively effective atdealing with low level online disputes (much like the medieval lawmerchant developed to resolve disputes quickly and practicallybetween people attending at markets and fairs from different landsspeaking different languages, similar to the informal"legal" systems developed by pirates in the 17th centuryto divide captured loot).

In British Columbia, the CRT (Civil Resolutions Tribunal) hasbeen introduced in an effort to relieve pressure on the highercourts by filtering out a significant volume of the smallerdisputes.

The CRT represents a free dispute resolution tool which requiresthe parties to engage in mediation at first instance, prior toprogressing to commence proceedings. Mediation fails in only areported two percent of cases. Litigated claims are almost entirelydealt with on the papers through the CRT, though the parties mayrequest an oral hearing via telephone or video link.

Other jurisdictions, including the UK, have been encouraged bythe success of the CRT and have pledged resources to developingapplicable ODR platforms to facilitate a much needed alleviation onthe stressed court system.

UK policy advocates have foreshadowed online dispute resolutionclauses in smart contracts, particularly in financial services.

Smart contracts are designed to enable transactions andagreements to be effected between disparate, often anonymousentities, without needing a central authority, or, presently; anexternal enforcement mechanism.

At the other end of the spectrum, a team of academics havedeveloped a bot fluent in AI that is apparently capable ofpredicting judgments made in the European Court of Human Rightswith a (fairly respectable) 79% accuracy rate.

Quite what the utility of this bot in the real world might beremains to be seen, but the benefit of any advanced form of AI isthat it intuits both its successes and failures and adaptsaccordingly via an algorithmic iterative learning process.

Legal AI, the Voldemort for some in the legal world, is likely away off from being implemented in Australian ODR.

However, there are promising applications of AI in certain areasof law, particularly those that involve a tangible prediction taskfor which there is a large volume of data available.

An example of this is the algorithm developed by Kleinberg etal, which was able to predict which bail applicants were likely tocommit a crime upon release. When the AI based application was setto the same release rate as the judges, the algorithm's choicesof applicants to release on bail committed 24.7% fewer crimes thanthose selected by the judges. The computer program only relied onthe defendants' age and charges in making its judgment, whilethe judges engaged with the defendants in open court.

It should be noted that AI based criminal sentencing is thesubject of significant human rights concerns in the UK and the USand it is unlikely that the Australian legal system will adopt it(especially in light of the political issues relating to apparentmistakes in the Australian government's AI based collection ofalleged overpaid welfare), but in commercial law things are likelyto be quite different.

The Federal Court of Australia has produced a machine learningconcept, reliant on AI, designed to aid parties to divide assetsand liabilities following a separation.

Such a division is usually debated at length (and cost) byfamily lawyers, however, the Federal Court is testing with what ithas named the "FCA Consent Order AI Application" toassist parties to determine a more accurate split likely to obtainthe Court's approval. The costs reductions inherent will likelymake this attractive to parties.

The current developments in AI are aimed at removing task-basedfunctions of the court and the judiciary, rather than replacing thejudicial function.

While we do not anticipate that we will be beholden to JudgeDredd any time in the immediate future, the digitisation of thelegal function has started to infiltrate our systems and the way inwhich we conduct dispute resolution. Litigants are not merelyobliged, but incentivised to adapt and engage with the newstructure.

Toby Blyth

Maddison Ives

Alternative dispute resolution

Colin Biggers & Paisley

The content of this article is intended to provide a generalguide to the subject matter. Specialist advice should be soughtabout your specific circumstances.

Here is the original post:
Artificial intelligence and the rise of online dispute resolution - Mondaq News Alerts

INSIDE THE FIRST AMENDMENT: ‘Getting it right’ – on Kobe Bryant and everything else – Meridian Star

Getting it right is one reliable defense for a free press in todays media world against critics who often base objections and critiques more on political differences than factual error.

However, criticism for getting it wrong is fair game for press skeptics with wrong covering a multitude of alleged sins, as occurred following the Jan. 26 deaths of NBA superstar Kobe Bryant, his daughter and seven others in a helicopter crash.

Even though many times the news outlets themselves quickly corrected errors or apologized, those moves often fell short of placating many on social media.

The BBC apologized quickly Sunday for using video of NBA star LeBron James during a segment on Bryants death a mistake the harkened to an old racist saw that to whites, all black people look alike.

The BBCs quick apology: In tonights coverage of the death of Kobe Bryant on #BBCNewsTen, we mistakenly used pictures of LeBron James in one section of the report, BBC Editor Paul Royall tweeted hours later. We apologize for this human error, which fell below our usual standards.

A slew of online critics some starting an online petition calling for resignation have questioned the sincerity of an MSNBC anchors apology Sunday after she appeared to use the n-word when reporting on Bryants death. Alison Morris later posted on Twitter: Earlier today, while reporting on the tragic news of Kobe Bryants passing, I unfortunately stuttered on air, combining the names of the Knicks and the Lakers to say Nakers. Please know I did not & would NEVER use a racist term. I apologize for the confusion this caused.

Gossip site TMZ was the first to report Bryants death. Los Angeles County Sheriff Alex Villanueva later said, It would be extremely disrespectful to understand that your loved one perished and you learn about it from TMZ. Los Angeles County Undersheriff Tim Murakami tweeted that he understood the pressures related to getting the scoop, but please allow us time to make personal notifications to their loved ones. Its very cold to hear of the loss via media.

Less clear is why The Washington Post placed a national political reporter, Felicia Sonmez, on administrative leave Monday only to reverse the action Tuesday because of tweets that began with a link to a 2016 Daily Beast story titled, Kobe Bryants Disturbing Rape Case: The DNA Evidence, the Accusers Story and the Half-Confession.

It is unclear whether the Posts initial action was in response to many online critics who called the tweet insensitive some issuing death threats, Sonmez said or because one of her tweets on the subject included a screen shot of her work email inbox, showing the names of critics.

By Tuesday, newsroom colleagues were rallying around Sonmez and Post columnist Erik Wemple wrote that the newspapers concerns, per an email from management to Sonmez, were that they didnt pertain to the reporters coverage area and that your behavior on social media is making it harder for others to do their work as Washington Post journalists. In the same column, Sonmez was quoted as saying she was never told the suspension involved the screen grab of her work email box.

Wemple raised questions about the action, noting that if journalists at the Post are prone to suspension for tweeting stories off their beats, the entire newsroom should be on administrative leave.

On Tuesday evening, After conducting an internal review, we have determined that, while we consider Felicias tweets ill-timed, she was not in clear and direct violation of our social media policy, Tracy Grant, managing editor of the Post, said in a statement.

Several news outlets drew a line from an early Fox News report with the incorrect number of those who died to a President Trump tweet repeating the error. ABC News on Wednesday suspended correspondent Matt Gutman after he erroneously reported on Sunday that all four of Bryants daughters were on the helicopter that crashed. Gutman, who also reported strong criticism on social media, apologized: We are in the business of holding people accountable, and I hold myself accountable for a terrible mistake, which I deeply regret.

A common factor here is the desire for a speedy post, reaction or comment, seemingly based on an assumption that readers and viewers and listeners care most about hearing news and seeing reactions now. But what about the values of accuracy, deliberation and thoroughness in an era in which much of the nation considers the news media unreliable, if not deliberately manipulative and unduly provocative?

Yes, many major news operations got it right by waiting for information to be verified. The Los Angeles Times at least took the intermediate approach of acknowledging online the early accounts of Bryants death and advising its readers it still was investigating those reports.

Sonmezs incident is a more challenging call for the Post, for the public and for a free press generally. Certainly, there is the longstanding social pressure and even journalistic guidelines to avoid sensationalism and inflicting undue pain that seems to lean toward avoiding disparaging news of a person immediately after his/her death.

But then, to ignore a womans allegation and resulting legal action in the #MeToo era (or anytime) likely would have resulted in legitimate complaints that journalists were channeling gender bias and hero worship and sanitizing a life. For the record, criminal charges against Bryan were dropped and a civil lawsuit settled out of court.

There is not one perfect way to gather or report news. The tasks are too complex and dependent on facts of each instance to produce cookie-cutter rules. And to large degree, the First Amendment rules out the enforceable codes of conduct that govern professions such as law and medicine.

But having been a young reporter at one time whose on-deadline job all too often was to collect a photo and interview family members of a recently deceased newsworthy person, I know there is a better way, with sensitivity, honesty and an emphasis on getting things right the first time, even when giving audiences an accurate news account.

Those terms sensitivity, honesty, getting it right and accurate are not in the First Amendments 45 words, to be sure. But that doesnt mean those values arent as real or applicable to how a free press should operate.

Gene Policinski is president and chief operating officer of the Freedom Forum Institute. He can be reached at gpolicinski@freedomforum.org, or follow him on Twitter at @genefac.

The rest is here:
INSIDE THE FIRST AMENDMENT: 'Getting it right' - on Kobe Bryant and everything else - Meridian Star