CLOSE
Original image
YouTube / IBM

How IBM Watson Learns

Original image
YouTube / IBM

Most of us know IBM's Watson computing system from its breakout performance on Jeopardy! a few years back; I covered that earlier today.

But Watson is significant not because it can win at Jeopardy! -- it's significant because it embodies a fundamental shift in how humans interact with computer systems. The new model is that we ask questions, Watson makes connections based on its ability to understand human language, and then it suggests possible answers...along with showing its work.

The fact that we can see the work behind Watson's answers is critically important -- that's not something you get from a simpler system like a search engine. Most interesting is that, because a human selects from among the best answers, humans can teach Watson in a positive feedback loop. Watson can even ask clarifying questions, allowing it to learn yet more about the world, and improving future performance. Watson works in tandem with humans, and sometimes the top answer is not the most useful to us -- it's the second, third, or fourth answer that may hold the key for a rare medical diagnosis, or an obscure connection. By deploying Watson in health care, IBM is helping doctors explore and improve medical care. Let's take a peek inside the IBM Watson Solutions Lab:

IBM calls Watson a "Learning System," and suggests it's the way we will interact with big data in the future. It's an intriguing notion, and it feels right to me. Especially when we talk about applications like health care, the ability for a human to help teach the computer is crucial. Got three and a half minutes to dig into how Watson learns? Watch this video.

If that piqued your interest, here's Manoj Saxena in a longer TEDx talk adding more context, including the tidbit that IBM is training Watson thoroughly enough that it's "within striking distance" of passing the U.S. Medical Licensing Exam (!):

I'll be digging deeper into Watson later today -- stay tuned for more Watson goodness.

Original image
iStock // Ekaterina Minaeva
arrow
technology
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
iStock
arrow
technology
Why Your iPhone Doesn't Always Show You the 'Decline Call' Button
Original image
iStock

When you get an incoming call to your iPhone, the options that light up your screen aren't always the same. Sometimes you have the option to decline a call, and sometimes you only see a slider that allows you to answer, without an option to send the caller straight to voicemail. Why the difference?

A while back, Business Insider tracked down the answer to this conundrum of modern communication, and the answer turns out to be fairly simple.

If you get a call while your phone is locked, you’ll see the "slide to answer" button. In order to decline the call, you have to double-tap the power button on the top of the phone.

If your phone is unlocked, however, the screen that appears during an incoming call is different. You’ll see the two buttons, "accept" or "decline."

Either way, you get the options to set a reminder to call that person back or to immediately send them a text message. ("Dad, stop calling me at work, it’s 9 a.m.!")

[h/t Business Insider]

SECTIONS
BIG QUESTIONS
arrow
BIG QUESTIONS
WEATHER WATCH
BE THE CHANGE
JOB SECRETS
QUIZZES
WORLD WAR 1
SMART SHOPPING
STONES, BONES, & WRECKS
#TBT
THE PRESIDENTS
WORDS
RETROBITUARIES