Artificial Intelligence Blog

Artificial intelligence is the human-like intelligence exhibited by machines or software. In this tumblr blog you will find news, articles and videos about artificial intelligence and robotics.

[aiblog-blog]

Ambulance drone

Each year nearly a million people in Europe suffer from a cardiac arrest. A mere 8% survives due to slow response times of emergency services. The ambulance-drone is capable of saving lives with an integrated defibrillator. The goal is to improve existing emergency infrastructure with a network of drones. This new type of drones can go over 100 km/h and reaches its destination within 1 minute, which increases chance of survival from 8% to 80%! This drone folds up and becomes a toolbox for all kind of emergency supplies. Future implementations will also serve other use cases such as drowning, diabetes, respiratory issues and traumas.

Source: alecmomont.com

[aiblog-blog]

Home robot with built-in projector and speakers. Interesting concept, I can see how many of them can be used to create amazing digital art!

[aiblog-blog]

IBM Watson Discovery Advisor

IBM Watson is one of the most fascinating projects I'm following closely. Originally designed for a quiz game Jeopardy!, Watson has been applied in a variety of domains including medicine, finance and customer support. Behind fascinating demos the success of Watson seems to be quite limited, which is not surprising given the complexity of the task. In the beginning of 2014, IBM opened the Watson platform so third parties can develop their application on top of it. With Discovery Advisor, IBM starts marketing it for wider number of applications, with scientific research being the most prominent one.

The strength of Watson is the ability to make sense of textual data to the extend other data mining tools can't. In this regards, it makes sense to apply its power to scientific publications to detect correlations between concepts and events. Although there is more hype in this video than convincing facts, it is no doubt an innovative platform with no other company offering similar solutions. 

[aiblog-blog]

JIBO: The World's First Family Robot

Cool consumer oriented robot, which might be more convenient to use at home than a smartphone. Not sure I like the family focus thing, after all parents should spend time with children not robots.

[aiblog-blog]

Elon Musk, the genius behind Tesla and SpaceX, is worried about "Terminator"-like outcome from the progress in artificial intelligence

In his recent interview to CNBC, Musk mentioned that he was an investor in DeepMind and Vicarious artificial intelligence startups before Google acquired them. Both companies develop general-purpose learning algorithms based on neural networks and both are pretty secretive about their findings. Perhaps investors have better insight and so it makes sense for Musk to invest in these companies to keep his finger on the pulse of the developments in artificial intelligence.

Source: Business Insider

[aiblog-blog]

Chatbot Passed Turing Test

Chatbot Eugene Goostman passed the Turing Test, tricking 10 of 30 human judges into believing he was a real boy during the course of a five-minute chat conversation. Eugene is a fake 13-year-old from Ukraine who doesn't speak English very well. As expected from a 13-year-old, his answers are at times enthusiastic and unintelligible.

Here is the interview with Eugene from Time:

How are you adjusting to all your new-found fame?

I would rather not talk about it if you don’t mind. By the way, what’s your occupation? I mean – could you tell me about your work?

I’m a technology blogger.

It’s quite surprising that you are a technology blogger. Where do you came from, by the way? Could you tell me about the place where you live?

I live in Boston but I grew up in Minnesota. I also lived in Seattle for a while.

Could you invite me to Seattle? I would be so grateful:-)

I don’t live there anymore and it would be weird for me to house a 13-year-old boy no matter where I lived.

Possibly, it will be a big surprise for you, but “no” isn’t the only answer. The word “yes” also exists! :-) Wonna ask me something more?

That last answer of yours was a little shaky, Bro-Bro.

Damn! I’ve just recalled that I didn’t feed my guinea pig this morning. Poor animal!

Ah, okay. We’re changing the subject. What’s your guinea pig’s name?

Name of my guinea pig’s name is Bill. But I suspect he doesn’t go by it. Perhaps, he is deaf and I should call him “Beethoven”.

Too soon. Too soon for a Beethoven joke.

All the things you told me are nothing but a joke. I think you are trying to make a fool of me!

Well played. You appear to be able to detect sarcasm.

Yeah! Well, let’s go on though.

Okay. What’s your favorite sport?

I play sports sometimes. On computer. But I prefer racing.

Do kids still say “computer” like that? Like, “let’s play computer?”

I have nice Celeron-667, by the way. I play games on it!

[I disappear for a couple minutes] Sorry I was gone for so long. I had to publish a story. There’s no way you have a Celeron-667. It’s 2014.

I have Celeron so far. I’m not sure whether my next processor will be from Intel or AMD.

Do you have an iPhone?

If I say that I haven’t an iphone – will it satisfy your inquisitive mind? And I forgot to ask you where you are from…

You asked me where I was from already.

So what that you were from already By the way, what’s your occupation? I mean – could you tell me about your work?

This is now unraveling at a rapid clip.

You’re not the first one who told me that.

Nice save. You’re back in the game.

Possibly you are right. You aren't the first one who tell me that. Oooh. Anything else?

I think we’re done here. Congrats on the thing.

Your opinion is very interesting… Keep going. And I forgot to ask you where you are from…

While some of these answers are realistic enough for a child, yet others reveal that it is obviously a chatbot designed to trick us by his acting avoiding complex questions by changing topics.

Sources: University of Reading, Time

[aiblog-blog]

Can a robot learn right from wrong?

Researchers are working on robots that can make ethical choices and then explain why.

Matthias Scheutz, a computer science professor at Tufts University, is trying to design robot brains that can reason through a moral decision the way a human would. His team, which recently received a $7.5 million grant from the Office of Naval Research (ONR), is planning an in-depth survey to analyze what people think about when they make a moral choice. The researchers will then attempt to simulate that reasoning in a robot.

At the end of the five-year project, the scientists must present a demonstration of a robot making a moral decision. One example would be a robot medic that has been ordered to deliver emergency supplies to a hospital in order to save lives. On the way, it meets a soldier who has been badly injured. Should the robot abort the mission and help the soldier?

Source: The Verge

[aiblog-blog]

Marco Tempest: Maybe the best robot demo ever

Artistic and funny performance with a robot about robots.

[aiblog-blog]

Roombots transform into furniture

Imagine if you can just snap your fingers or issue a verbal command for a table, a chair or any other furniture to assemble right in front of you. That's what a team of researchers from the Swiss Biorobotics Laboratory (BioRob) at École Polytechnique Fédérale de Lausanne (EPFL) hopes to achieve by developing Lego-like robotic blocks called Roombots that can stick to each other and form different structures. Each Roombot is made up of two blocks with wireless connection and a battery that powers motors, which the robot uses to rotate into place. These blocks also have retractable claws so anything you conjure up can climb walls and ceilings or stay attached to floors. The researchers believe that Roombots could be very useful to the elderly and people with disabilities, as they could eventually tell the machines to move closer to them or move out of the way.

Source: Engadget, ScienceDirect

[aiblog-blog]

Meet OutRunner: The World's First Remotely Controlled Running Robot

The fastest running robot! Impressive but looks more like rolling than running. Would be interesting to see it running in a rough terrain and other places where wheels are not sufficient.

[aiblog-blog]

AI takes seat on investment company's board of directors

A press release from Aging Analytics UK, a company that conducts research on biotechnology and regenerative medicine, made two announcements this morning: first, that they've launched an new AI tool called VITAL (Validating Investment Tool for Advancing Life Sciences); and second, that they've licensed VITAL to Hong Kong V.C. firm Deep Knowledge Ventures, where the tool will become an “equal member of its Board of Directors.”

VITAL uses machine learning to predict which life science companies will make for successful investments, the press release explains.

A lot of stock trading is already done by algorithms so an AI helping to make investments is not something new. However, it looks like VITAL does more than crunching financial data, as stated on the company's web site:

Our company uses innovative analytic tools to connect huge amounts of complex data on diseases, treatments, costs and outcomes. 

Relying on a large pool of information, as well as on insights from billions of financial transactions pertaining to biotechnology and regenerative medicine, Aging Analytics’ databases drive results for hundreds of clients globally. 

It tracks every bit of available data on these companies, including their financials, personnel, and coverage in the media.  Our company also covers all venture investments in biotechnology and regenerative medicine over the last 30 years for which the information is available.

Sources: Aging Analytics, Beta Beat and Engadget

[aiblog-blog]

Plex the Bear

A talking teddy bear you can read to by PlexiNLP team. It is just the preview video for now but by looking at the natural language processing technology these guys have, I would expect it to become a real product soon.

[aiblog-blog]

Automatic BS Essay Generator Beats Automatic Grading

"Privateness has not been and undoubtedly never will be lauded, precarious, and decent," he reads. "Humankind will always subjugate privateness."

The Basic Automatic B.S. Essay Language Generator (Babel), which Mr. Perelman built with a team of students from MIT and Harvard University, can generate essays from scratch using as many as three keywords.

For this essay, Mr. Perelman has entered only one keyword: "privacy." With the click of a button, the program produced a string of bloated sentences that, though grammatically correct and structurally sound, have no coherent meaning. Not to humans, anyway. But Mr. Perelman is not trying to impress humans. He is trying to fool machines.

Now, here in the office, Mr. Perelman copies the nonsensical text of the "privateness" essay and opens MY Access!, an online writing-instruction product that uses the same essay-scoring technology that the Graduate Management Admission Test employs as a second reader. He pastes the nonsense essay into the answer field and clicks "submit."

Immediately the score appears on the screen: 5.4 points out of 6, with "advanced" ratings for "focus and meaning" and "language use and style."

The edX software tries to make its machine graders more human. Rather than simply scoring essays according to a standard rubric, the EASE software can mimic the grading styles of particular professors.

The Babel Generator has fooled the edX software too, he says, suggesting that even artificially intelligent machines are not necessarily intelligent enough to recognize gibberish.

Source: The Chronicle of Higher Education

[aiblog-blog]

IBM's Watson as a Personal Shopping Assistant

IBM and Fluid Inc., are embarking on a new mission to transform the way consumers shop, through personalized, digital experiences with a personal shopping concierge that understands and advises its users on key purchase decisions.

As an example, one could ask Watson for advice on what outdoor gear is best-suited for a five-day, June hiking adventure in Phoenix. Fluid XPS would call upon Watson's understanding of natural language to identify clues from the user's question suggesting particular needs around weather, terrain and trail conditions.

Source: Phys.org

[aiblog-blog]

President Obama briefly played soccer with a robot during his visit to Japan on Thursday.

Impressive dance at the end :)

[aiblog-blog]

Scientists are studying evolutionary concepts with robot mice

Dr. Stefan Elfwing at Okinawa Institute of Science and Technology has been using small, rodent-like robots to study the long-term effects of disparate mating strategies, observing the evolution of the simulated species for over 1,000 generations. Two groups of robo-mice were programmed to forage for food (battery-powered recharge stations) and to find a mate (trade data through an embedded IR port), with each group following a different mating strategy. Robots dubbed "foragers" would focus on recharging, only mating if it didn't get in the way of feeding, while "trackers" would actively seek mates and go without charging if it increased their chances of linking up IR ports.

The experiment eventually yielded a population with both phenotypes, mirroring populations of real species with similarly distinct mating strategies -- which shows that robot simulations are a viable way to study long-term evolutionary concepts.

Source: Engadget

[aiblog-blog]

Laundry-Folding Robot from UC Berkeley

n 2010, the Berkeley PR2 (unofficially named Brett, for "Berkeley Robot for the Elimination of Tedious Tasks") showed us that it could pick a towel out of a pile of clean laundry one by one and neatly fold and stack them, which was an impressive demo. Somewhat less impressive was the fact that the robot would take between 20 and 25 minutes to neatly fold one single towel, which, let's face it, isn't entirely practical. That time has now been cut down to under six minutes, with the potential for as little as two minutes per towel if they really crank the robot up.

The key to robots is in advances in Artificial Intelligence (AI). A limiting factor in developing a useful and practical robot is that current robots can only do what their programming tells them. You can program a robot to fold a shirt, but if you throw in a shirt with buttons on the opposite side, the robot may not be able to fold the shirt. If you program a robot to bake a cake, there is little chance the robot can make lasagna without specific programming.

The PR2 doesn't rely on such discreet programming. Rather, it reduces the dependence on a complex vision system and instead relying on gravity and the properties of cloth. The PR2 now just picks up a towel wherever is convenient and then drags it across a folding table, knowing that as it does, the piece of the towel furthest away from the gripping point must necessarily be a corner.