Google Hor

advertisement

Tech News

Playkey raises $2.8M to fund its US expansion

Playkey raises $2.8M to fund its US expansion

Playkey, a game streaming service we saw earlier this year at Disrupt NY, has just closed an additional $2.8 million round of funding from Russia’s

...

Playbuzz unveils a new editor for writing articles chock full of interactive content

SF city attorney seeks court order to force Lyft and Uber to hand over driver data

SF city attorney seeks court order to force Lyft and Uber to hand over driver data

San Francisco City Attorney Dennis Herrera

...

Why the future of deep learning depends on finding good data

Ophir Tanz is the CEO of GumGum , an artificial intelligence company with particular expertise in computer vision. GumGum applies its capabilities to a variety of industries, from advertising to professional sports across the globe. Ophir holds a B.S. and a M.S. from Carnegie Mellon University and currently lives in Los Angeles.

More posts by this contributor:

Cambron Carter Contributor

Cambron Carter leads the image technology team at GumGum , where he designs computer vision and machine learning solutions for a wide variety of applications. Cambron holds B.S. degrees in physics and electrical engineering and an M.Eng. in electrical engineering from the University of Louisville.

More posts by this contributor:

We’ve already taken a look at neural networks and deep learning techniques in a previous post , so now it’s time to address another major component of deep learning: data—meaning the images, videos, emails, driving patterns, phrases, objects, and so on that are used to train neural networks.

Surprisingly, despite our world being quite literally deluged by data—currently about 2.5 quintillion bytes a day, for those keeping tabs—a good chunk of it is not labeled or structured, meaning that for most current forms of supervised learning it’s unusable. And deep learning in particular depends on a steady supply of the good, structured and labeled stuff.

In the second part of our “A Mathless Guide to Neural Networks”, we’ll take a look at why high-quality, labeled data is so important, where it comes from, how it’s used, and what solutions our eager-to-learn machines can expect in the near-term future.

Supervised learning: I wanna hold your hand

In our post about neural networks, we explained how data is fed to machines through an elaborate sausage press that dissects, analyzes, and even refines itself on the fly. This process is considered supervised learning in that the giant piles of data fed to the machines have been painstakingly labeled in advance. For example, to train a neural network to identify pictures of apples or oranges, it needs to be fed images that are labeled as such. The idea is that machines can be groomed to understand data by finding what all pictures labeled apple or orange, respectively, have in common, so they can eventually use those recognized patterns to more accurately predict what they are seeing in new images. The more labeled pictures they see, the bigger (and more diverse) the dataset, the better they can refine the accuracy of their predictions; practice makes (almost) perfect.

This approach is useful in teaching machines about visual data, and how to identify anything from photographs and video to graphics and handwriting. The obvious upside is that it is now relatively commonplace for machines to be equal or even better than humans at say, image recognition for a number of applications. For instance, Facebook’s Deep Learning software is able to match two images of an unfamiliar person at the same level of accuracy as a human   (better than 97% of the time), and Google, earlier

...

Sami Atiya from ABB says industrial robots will add jobs, not take them away

Sami Atiya from ABB says industrial robots will add jobs, not take them away

In and interview earlier this week at the  TechCrunch Robotics Session  held on the MIT campus in Cambridge, MA, Sami Atiya, president of ABB’s Robotics and Motion division, said he believes bringing robots into the manufacturing process actually adds jobs instead of killing them.

ABB certainly has some data points with over 300,000 industrial robots installed worldwide, and Atiya claims that conventional wisdom is wrong when it comes to robots and jobs. “Automation is going to drive more productivity and also jobs,” he said. He went onto say that countries with the highest ratios of humans to industrial robots in production environments, also have the lowest rates of manufacturing unemployment.

“If you look at pure data and statistics,” he said, “in the countries that have the highest rates of robots per employees, which is Japan and Germany, they have about 300 robots per 10,000 employees, and they have the least unemployment…in the manufacturing sector.”

He also claimed that there have been 100,000 industrial robots installed in the US in the last five years, which has resulted in 270,000 additional jobs, more than two jobs for every robot. (ABB cites the International Federation of Robotics, World Bank, OECD and BLS as sources for these numbers.)

There has been of course lots of speculation that as companies increase the use of robots to automate jobs, that there will be a corresponding job loss. In May, an article in the the LA Times appeared to back up this assertion, citing a study by PwC, which  claimed that 38 percent of all US jobs could be lost to automation by the early 2030s. That’s a frightening prospect to many people and to policy makers who would have to deal with the fallout if that were to happen.

An article on CNN Money from last March, smack dab in the middle of the contentious presidential campaign, cited numbers from the Bureau of Labor Statistics that 5 million manufacturing jobs have been lost since 2000. While there has been much debate for the reason, the article claims robots and machines have been a big contributing factor in replacing workers. It’s worth noting that there are still over 12 million jobs in the sector in spite of decades of steady decline.

Sami Atiya from ABB says industrial robots will add jobs, not take them away

ABB Robot arm. Photo: Veanne Cao, TechCrunch

Atiya said one of the reasons companies are moving to robots is they simply can’t compete without them. “If you look at this from a macro-[economic] perspective, skilled labor is becoming [more scarce], and it’s not a question [whether] you want to do it or not. You have to do it to stay competitive as a nation, and also as a company,” he said.

Atiya used the standard argument for these types of historical economic transitions comparing the increasing use of robotics with the rise of the steam engine, electricity and industrialization. The common belief during all of these key changes was that they would kill jobs, but in the end they created more jobs because of productivity increases,

...