google AI predicts heart attacks by scanning your eye…

This week, the geniuses at Google and its “health-tech subsidiary” Verily announced AI that can predict your risk of a major cardiac event with roughly the same accuracy as the currently-accepted method using just a scan of your eye.

They have created an algorithm that analyzes the back of your eye for important predictors of cardiovascular health “including age, blood pressure, and whether or not [you] smoke” to assess your risk.

As explained via The Verge:

“To train the algorithm, Google and Verily’s scientists used machine learning to analyze a medical dataset of nearly 300,000 patients. This information included eye scans as well as general medical data. As with all deep learning analysis, neural networks were then used to mine this information for patterns, learning to associate telltale signs in the eye scans with the metrics needed to predict cardiovascular risk (e.g., age and blood pressure).

When presented with retinal images of two patients, one of whom suffered a cardiovascular event in the following five years, and one of whom did not, Google’s algorithm was able to tell which was which 70 percent of the time. This is only slightly worse than the commonly used SCORE method of predicting cardiovascular risk, which requires a blood test and makes correct predictions in the same test 72 percent of the time.

Why It’s Hot:

This type of application of AI can help doctors quickly know what to look into, and shows how AI could help them spend less time diagnosing, and more time treating. It’s a long way from being completely flawless right now, but in the future, we might see an AI-powered robot instead of a nurse before we see the doctor.

[Source]

YouTube’s Algorithm Helps Spread Lies

Time to re-litigate our favorite Hot Sauce topic! Are social media platforms media platforms or not? Do they have a journalistic responsibility to the public?

Last week YouTube’s quickly algorithm spread a rumor that Parkland HS shooting survivor David Hogg was an actor hired by democrats. It had 200K views shortly after it’s posting, but has since been taken down.

Mashable, the source for this post had a very apt understanding of the problem at hand:

YouTube’s job, as it sees it, is to get as many eyeballs on as many videos as possible. It’s as if a media tycoon founded a newspaper, invited every conspiracy theorist to contribute, and blithely waved away the notion that there should be any ethical responsibility to put forth the verifiable truth — because selling ads was all that mattered.

In a recent Guardian study this recent Guardian study out of 643 of partisan videos  recommended to people watching politics content in 2016, 551 were conspiracy-based content that favored Trump while 92 favored Clinton. This study also notes that “More than 80% of the YouTube-recommended videos about the pope detected by his program described the Catholic leader as “evil”, “satanic”, or “the anti-Christ”.”

The Guardian tells the story of french programmer Guillaume Chaslot founder of https://algotransparency.org/ who was looking to make changes internally at YouTube in 2013 before he was fired (reportedly for performance reasons, but I’ll leave that up to you.) He believes that the YouTube Algorithm was biased towards Trump because it is biased towards divisiveness and conspiracy (things that shock and awe audiences can get more clicks). You can read his full blog post on the subject here.

Why Its Hot?

Let’s continue the conversation about how AI and algorithms shape the way we live, is there a way we can make them more human instead of human pandering.

the camera doesn’t lie, but the algorithm might…

Algorithms fooling algorithms may be one of the most 21st century things to happen yet. But, it did. Researchers at MIT used an algorithm to 3D print versions of a model object, programming them to be recognized as certain other things by Google’s image recognition technology. In short, they fooled Google image recognition into thinking a 3D printed stuffed turtle was a rifle. They also made a 3D printed stuffed baseball appear to be espresso, and a picture of a cat appear to be guacamole. Technology truly is magic.

Their explanation:

“We do this using a new algorithm for reliably producing adversarial examples that cause targeted misclassification under transformations like blur, rotation, zoom, or translation, and we use it to generate both 2D printouts and 3D models that fool a standard neural network at any angle.

It’s actually not just that they’re avoiding correct categorization — they’re classified as a chosen adversarial class, so we could have turned them into anything else if we had wanted to. The rifle and espresso classes were chosen uniformly at random.”

Why it’s hot:
Clearly there are implications for the practicality of image recognition. If they can do this fairly easily in a lab setting, what’s to stop anyone with enough technical savvy from doing this in the real world, perhaps reversing the case and disguising a rifle as a stuffed turtle to get through an artificially intelligent, image recognition technology-driven security checkpoint? Another scary implication mentioned was self-driving cars. It just shows we need much more ethical hacking to plan for and prevent these kind of security concerns.

Disney World Crowd Algorithms, Retrofitted for MedTech

Until recently, Len Testa used his computer science training to help people optimize their trips to Disney World through his algorithm-powered company Touring Plans. Now, he’s using his algorithms for something totally different: clinical diabetes decision-making software.

Recently approved by the FDA for clinical trial, GlucosePATH is an application that processes multiple data points and decides which medicine(s) their physicians should probably prescribe. GlucosePATH is significant because, in addition to processing medical data points, it incorporates the patient’s insurance data and factors financial considerations into its final decisions. (What good is a carefully chosen medicine if the patient can’t afford it anyway?)

The programming portion of building GlucosePATH was startlingly simple, says Testa. “There are around 6 million different combinations of diabetes medications to choose from in a typical office visit, but there are 2,432,902,008,176,640,000 different ways to visit 20 rides in a theme park. So the theme park problem is about 400 billion times bigger in terms of things that have to be considered.” Testa says he’s surprised that more medical decision-making isn’t automated; his work with GlucosePATH may just clear the way.

Why it’s hot: Tourism and medtech seem like they wouldn’t have much in common, but this teaming illustrates just how wrong that assumption is. This combination is a fantastic example of the possibilities of cross-pollenated ideas, no matter where they come from. What other solutions can we find by looking in unexpected places?

Learn more: MedGadget

googler creates AI that creates video using one image…

One of the brilliant minds at Google has developed an algorithm that can (and has) create video from a single image. The AI does this by predicting what each of the next frames would be based on the previous one, and in this instance did it 100,000 times to produce the 56 minute long video you see above. Per its creator:

“I used videos recorded from trains windows, with landscapes that moves from right to left and trained a Machine Learning (ML) algorithm with it. What you see at the beginning is what the algorithm produced after very little learnings. It learns more and more during the video, that’s why there are more and more realistic details. Learnings is updated every 20s. The results are low resolution, blurry, and not realistic most of the time. But it resonates with the feeling I have when I travel in a train. It means that the algorithm learned the patterns needed to create this feeling. Unlike classical computer generated content, these patterns are not chosen or written by a software engineer.

Why it’s hot:

Creativity and imagination have been among the most inimitable human qualities since forever. And anyone who’s ever created anything remotely artistic will tell you inspiration isn’t as easy as hitting ‘go’. While this demonstration looks more like something you’d see presented as an art school video project than a timeless social commentary regaled in a museum, it made me wonder – what if bots created art? Would artists compete with them? Would they give up their pursuit because bots can create at the touch of a button? Would this spawn a whole new area of human creativity out of the emotion of having your work held up next to programmatic art? Could artificial intelligence ever create something held up against real human creativity?

Going Digital to Rescue Food

About 40 percent of all food in America is wasted. Much of it ends up in landfills, where it emits dangerous-to-the-planet methane gas. At the same time, one out of eight American households don’t have enough to eat.

Wasted food. Hungry people. How do we get the two to meet?

Last month, food rescue made a leap to a national scale. Feeding America matches donors and recipients with an algorithm. A restaurant can go on Meal Connect to post an offer of, say, eight trays of fried chicken and biscuits. Meal Connect will automatically match that offer with the closest food pantry or soup kitchen that can get it up right away.

Feeding America

Meal Connect makes it possible to rescue prepared food and smaller quantities of food — and to do so quickly. “This allows us to provide real hot meals — virtually at the same time that someone coming off the street and paying for it would get it,” she said.

https://www.nytimes.com/2017/05/02/opinion/going-digital-to-rescue-food.html?smprod=nytcore-iphone&smid=nytcore-iphone-share&_r=0

Why It’s Hot
When the scale and reach of digital is brought to bear on social issues, huge problems become more surmountable.

Your Kid’s Computer Has Dinner Covered.

Neural networks are computer learning algorithms that mimic the interconnected neurons of a living brain, managing astonishing feats of image classification, speech recognition, or music generation by forming connections between simulated neurons.

I’m not a neural network expert, so I had to look that one up when I heard that there was a grad student who loaded a neural network code on her 2010 MacBook Pro, and started training it on a bunch of recipes and cocktails.

Here’s a few recipes the network has generated:

Pears Or To Garnestmeam

meats

¼ lb bones or fresh bread; optional
½ cup flour
1 teaspoon vinegar
¼ teaspoon lime juice
2  eggs

Brown salmon in oil. Add creamed meat and another deep mixture.

Discard filets. Discard head and turn into a nonstick spice. Pour 4 eggs onto clean a thin fat to sink halves.

Brush each with roast and refrigerate.  Lay tart in deep baking dish in chipec sweet body; cut oof with crosswise and onions.  Remove peas and place in a 4-dgg serving. Cover lightly with plastic wrap.  Chill in refrigerator until casseroles are tender and ridges done.  Serve immediately in sugar may be added 2 handles overginger or with boiling water until very cracker pudding is hot.

Yield: 4 servings

This is from a network that’s been trained for a relatively long time – starting from a complete unawareness of whether it’s looking at prose or code, English or Spanish, etc, it’s already got a lot of the vocabulary and structure worked out.

This is particularly impressive given that it has the memory of a goldfish – it can only analyze 65 characters at a time, so by the time it begins the instructions, the recipe title has already passed out of its memory, and it has to guess what it’s making. It knows, though, to start by browning meat, to cover with plastic wrap before chilling in the refrigerator, and to finish by serving the dish.

Compare that to a recipe generated by a much earlier version of the network:

Immediately Cares, Heavy Mim

upe, chips

3  dill loasted substetcant
1  cubed chopped  whipped cream
3  unpreased, stock; prepared; in season
1  oil
3 cup milk
1 ½ cup mOyzanel chopped
½ teaspoon lemon juice
1 ¼ teaspoon chili powder
2 tablespoon dijon stem – minced
30  dates afrester beater remaining

Bake until juice. Brush from the potato sauce: Lightly butter into the viscin. Cook combine water. Source: 0 25 seconds; transfer a madiun in orenge cinnamon with electres if the based, make drained off tala whili; or chicken to well. Sprinkle over skin greased with a boiling bowl.  Toast the bread spritkries.

Yield: 6 servings

which bakes first, has the source in the middle of the recipe directions, mixes sweet and savory, and doesn’t yet know that you can’t cube or chop whipped cream.

An even earlier version of the network hasn’t yet figured out how long an ingredients list should be; it just generates ingredients for pages and pages:

Tued Bick Car

apies

2 1/5 cup tomato whene intte
1 cup with (17 g cas pans or
½ cup simmer powder in patsorwe ½ tablespoon chansed in
1 ½ cup nunabes baste flour fite (115 leclic
2 tablespown bread to
¼ cup 12″. oz mice
1  egg barte, chopped shrild end
2 cup olasto hote
¼ cup fite saucepon; peppen; cut defold
12 cup mestsentoly speeded boilly,, ( Hone
1  Live breseed
1  22 ozcugarlic
1 cup from woth a soup
4 teaspoon vinegar
2 9/2 tablespoon pepper garlic
2 tablespoon deatt

And here’s where it started out after only a few tens of iterations:

ooi eb d1ec Nahelrs  egv eael
ns   hi  es itmyer
aceneyom aelse aatrol a
ho i nr  do base
e2
o cm raipre l1o/r Sp degeedB
twis  e ee s vh nean  ios  iwr vp  e
sase
pt e
i2h8
ePst   e na drea d epaesop
ee4seea .n anlp
o s1c1p  ,  e   tlsd
4upeehe
lwcc   eeta  p ri  bgl as eumilrt

Even this shows some progress compared to the random ASCII characters it started with – it’s already figured out that lower case letters predominate, and that there are lots of line breaks. Pretty impressive!

Why It’s Hot:
Progress, progress, progress. Sometimes we take for granted how long and arduous the road to further our convenience is, or how well-equipped technology actually gets us from point A to B. We don’t always need to look under that hood, but we should be happy someone does, and technology such as machine learning neural networks continue to evolve to make our lives easier-or more entertaining until they get something right.  As the ability to learn from the tons of content mankind has already created continues to improve, there really is some scary (don’t)DIY frontiers on the horizon. Forget about wondering if your kid lifted their essay content from an online wiki source, worry instead if he loaded a code, taught his Mac to ingest thousands of volumes of American history and spit out a dissertation on the significance of the Lincoln-Douglass debates without penning a word. Then don’t punish that kid, get him a job making me new cocktails.
Click here if you want to see the cocktails it created:

The Amazon “stock market”

Just beneath the placid surface of a typical product page on Amazon lies an unseen world, a system where third-party vendors can sell products alongside Amazon’s own goods. It’s like a stock market, complete with day traders, code-slinging quants, artificial intelligence algorithms and, yes, flash crashes.

Amazon

Sellers of commodity items on Amazon are constantly monitoring and updating their prices, sometimes hundreds of thousands of times a day across thousands of items, says Mr. Kaziuk nas. Most use “rules-based” pricing systems, which simply seek to match competitors’ prices or beat them by some small fraction. If those systems get into bidding wars, items offered by only a few sellers can suffer sudden price collapses — “flash crashes.”

It’s clear, after talking to sellers and the software companies that empower them, that the biggest of these vendors are growing into sophisticated retailers in their own right. The top few hundred use pricing algorithms to battle with one another for the coveted “Buy Box,” which designates the default seller of an item. It’s the Amazon equivalent of a No. 1 ranking on Google search, and a tremendous driver of sales.

http://news.morningstar.com/all/dow-jones/us-markets/20170326515/the-high-speed-trading-behind-your-amazon-purchase.aspx

Why It’s Hot

Getting under the hood of how retail monster Amazon operates is always fascinating. The idea that prices are updating hundreds of thousands of times a day is nuts.

When programmatic goes wrong

Jaguar Land Rover has ceased all digital advertising in the UK after an investigation revealed it was funding terror organizations without its knowledge.

Jaguar

According to reports the car marque’s programmatic ads were among a number of brands indirectly paying Islamic extremists, white supremacists and pornographers.

Ads for the Jaguar F-Pace have appeared on YouTube next to a pro-Isis video that has been viewed more than 115,000 times. It has since been removed.

In a statement Jaguar said: “Jaguar Land Rover is very concerned by reports that advertising featuring our brands may benefit extremist and other inappropriate on-line media. This is an unintended consequence of algorithm technology used on some video-sharing websites.

http://www.thedrum.com/news/2017/02/12/jaguar-land-rover-pulls-ads-amid-terror-funding-investigations

Why It’s Hot
-It reminds us that algorithms and quant data are not everything. Our ability to monitor, evaluate and draw learnings is key
-It also is a reminder of our responsibility, liability and yes sometimes vulnerability in engagements with our clients. I can imagine the phone calls Jaguar’s UK agency received. We need to be able to justify our decisions but also deal with our mistakes when they happen (and they will happen. and we will have each other’s backs!)

Algorithms For Sale

Seattle-based start-up Algorithmia is a new marketplace for algorithms. The company wants to make it easier for computer researchers to monetize the algorithms they create and publish in academic papers and make them available to developers/businesses who want to take advantage of the new technology. The system and marketplace will make it easy for prospective buyers to select the algorithm they want and then easily implement it into their applications with the use of the Algorithmia API.

Why It’s Hot:

 The majority of software products and applications are made up of highly complex algorithmsThe most famous algorithm led to the development of the $400 billion dollar giant Google, which has changed the world forever. As more technologies emerge and grow, and data continues to be produced at a rapid pace, faster hardware will no longer be able to support the architecture for data analysis. It will then become critical that organizations, people, and products focus on data analysis packages comprised of algorithms versus the actual hardware they run on.

 

A “Sick” New App

A new app called HEALTHYDAY, developed by Johnson & Johnson, uses sickness-searching algorithms to warn you when sickness in lurking in your neighborhood. The app’s algorithm syncs self-reported data form local doctor’s offices, Google searches, social media mentions on Twitter & Facebook and user data from people using the app, which it funnels down into easily digestible trends, blurbs and infographics. The goal of the app is to answer the question “What’s going around?”. Learn more about the app here.

Capture

 

Why It’s Hot:
This app serves as a best-in-class example of using “big data” to create something educates & addresses a need, while subtly pushing core products in the Johnson & Johnson family of brands.

Yahoo Takes The Scenic Route

Yahoo Labs is adding a dimension to their navigation algorithm that aims to provide more beautiful ways to get you to your chosen destination. Researchers at Yahoo Labs sourced images from Google Streetview and crowd-sourced user opinions about which streets were more beautiful via UrbanGems.org, then assigned an attractiveness variable to each location.

“The goal of this work is to automatically suggest routes that are not only short but also emotionally pleasant,” said Daniele Quercia, of Yahoo Labs. “Based on a quantitative validation, we find that, compared to the shortest routes, the recommended ones add just a few extra walking minutes and are indeed perceived to be more beautiful, quiet and happy.”

via Wired

On-the-road

Why It’s Hot?

Marketers often ask themselves how they can more effectively inject an emotional element to the work in order to get closer to what people actually feel and react to. This is a great example of taking a perfectly good pre-existing tool and adding a layer that can expose an element of beauty in what has become a very mundane and predictable experience.  This helps elevate technology to be more humanistic and useful in how it maps to real life travel behaviors.