LinkedIn Open-Sources FeatureFu, A Toolkit For Building Machine Learning Models

LinkedIn today announced that it is open-sourcing an internal tool called FeatureFu. The FeatureFu toolkit is meant to make it easier for developers to build their machine learning models around statistical modeling and decision engines.

The idea here is to take LinkedIn’s knowledge around “feature engineering” and make it accessible to developers outside of the company. In machine learning, feature engineering is basically using your detailed knowledge of the phenomenon you are looking at and then using that to build machine learning models.

LinkedIn argues that most large-scale recommendation systems (think LinkedIn’s own tools for suggesting connections on its site) are managed by at least two teams: one that handles the offline modeling and one that takes care of the online feature-serving/model-scoring part of the system. This leads to a number of problems FeatureFu is trying to solve.

“Many large-scale recommendation systems are brittle and vulnerable. FeatureFu allows for creative and agile development on these systems so that shipping new features doesn’t take weeks or even months,” LinkedIn senior software engineer Bing Zhao told me.

A small change in how features are generated can create lots of work for the other team, for example, and also makes it hard to experiment with different feature/model techniques.

FeatureFu uses a small Java library called Expr that developers can use to transform and build upon an existing set of features. “Once deployed to an online feature generation framework, it eliminates any further need for code-change to ship models for a wide range of derived features,” as Zhao describes the advantages of this system.

So why did LinkedIn decide to open source this tool? “When we have a business need for software, we first look to see if there are pre-existing software projects in open source. If there isn’t then we create it ourselves,” Zhao told me. “As long as the software isn’t a business differentiator, then we often open source the project so that many can benefit.”

Zhao also noted that he hopes for FeatureFu to become widely adopted. “FeatureFu could become a common technique for many machine learning systems,” he said. “It enables feature engineering to be more agile, which is one of the keys to success for machine learning applications. So we wanted to share our work with the industry.”

Comments

Best Web Design Company in Pondicherry

#Technology has two faces. We all feel it, but sometimes can’t find words to describe it. #Ebooks are the best example to show the 0-1 nature of emotions the #technology evokes. #itwhere provide a #Best #solutions to #Growyourbusiness feel free to drop a #Mail info@itwheretech.co.in www.itwheretech.co.in

How ad-free subscriptions could solve Facebook

At the core of Facebook’s “well-being” problem is that its business is directly coupled with total time spent on its apps. The more hours you pass on the social network, the more ads you see and click, the more money it earns. That puts its plan to make using Facebook healthier at odds with its finances, restricting how far it’s willing to go to protect us from the harms of over use. The advertising-supported model comes with some big benefits, though. Facebook CEO Mark Zuckerberg has repeatedly said that “We will always keep Facebook a free service for everyone.” Ads lets Facebook remain free for those who don’t want to pay, and more importantly, for those around the world who couldn’t afford to. Ads pay for Facebook to keep the lights on, research and develop new technologies, and profit handsomely in a way that attracts top talent and further investment. More affluent users with more buying power in markets like the US, UK, and Canada command higher ad prices, effectively

South Korea aims for startup gold

Back in 2011, when South Korea won its longshot bid to host the 2018 Winter Olympics, the country wasn’t widely recognized as a destination for ski and snow lovers. It wasn’t considered much of a tech startup hub either. Fast forward seven years and a lot has changed. For the next 10 days, the eyes of the world will be on the snowy slopes of PyeongChang. Meanwhile, a couple of hours away in Seoul, a burgeoning startup scene is seeing investments multiply, generating exits and even creating a unicorn or two. While South Korea doesn’t get a perfect score as a startup innovation hub, it has established itself as a serious contender. More than half a billion dollars annually has gone to seed through late-stage funding rounds for the past few years. During that time, at least two companies, e-commerce company Coupang and mobile-focused content and commerce company Yello Mobile, have established multi-billion-dollar valuations. To provide a broader picture of how South Korea stacks

Phoenix OS is (another) Android-as-a-desktop

Google Android may have been developed as a smartphone operating system (and later ported to tablets, TVs, watches, and other platforms), but over the past few years we’ve seen a number of attempts to turn it into a desktop operating system. One of the most successful has been Remix OS , which gives Android a taskbar, start menu, and an excellent window management system. The Remix OS team has also generated a lot of buzz over the past year, and this week the operating system gained a lot of new alpha testers thanks to a downloadable version of Remix OS that you can run on many recent desktop or notebook computers. But Remix OS isn’t the only game in town. Phoenix OS is another Android-as-desktop operating system, and while it’s still pretty rough around the edges, there are a few features that could make it a better option for some testers. Some background I first discovered Phoenix OS from a post in the Remix OS Google Group , although I’ve also found mentions of th

What will a driverless future actually look like?

There is a growing consensus that autonomous vehicles (AVs) will soon be a reality. The debate today centers not on whether, but how soon, AVs will be commonplace on our roads. But for all the buzz surrounding AVs, many details about what a driverless future will look like remain unclear. Which business models will work best for the commercialization of AVs? Which AV usage models will be most appealing for consumers? Which companies are best positioned to win in this new market? These are big questions, and no certain answers can be given at this stage. Nonetheless, it is valuable to reflect, in a concrete way, on how this transformative technology might develop. This article will present some conjectures. The end of private car ownership? At a high level, two possible paradigms seem most likely for how society will use AVs. The first is private AV ownership. Under this model, individuals or families would continue to own their own vehicles and use them to get around. As

Exciting Technology

Search This Blog