- `RankBrain' uses artificial intelligence to filter results
- In tests, system beats company's experts at page selection.
When Google-parent Alphabet Inc. reported eye-popping earnings last week its executives couldn’t stop talking up the company’s investments in machine learning and artificial intelligence.
For any other company that would be a wonky distraction from its core business. At Google, the two are intertwined. Artificial intelligence sits at the extreme end of machine learning, which sees people create software that can learn about the world. Google has been one of the biggest corporate sponsors of AI, and has invested heavily in it for videos, speech, translation and, recently, search.
For the past few months, a “very large fraction” of the millions of queries a second that people type into the company’s search engine have been interpreted by an artificial intelligence system, nicknamed RankBrain, said Greg Corrado, a senior research scientist with the company, outlining for the first time the emerging role of AI in search.
RankBrain uses artificial intelligence to embed vast amounts of written language into mathematical entities -- called vectors -- that the computer can understand. If RankBrain sees a word or phrase it isn’t familiar with, the machine can make a guess as to what words or phrases might have a similar meaning and filter the result accordingly, making it more effective at handling never-before-seen search queries.
The system helps Mountain View, California-based Google deal with the 15 percent of queries a day it gets which its systems have never seen before, he said. For example, it’s adept at dealing with ambiguous queries, like, “What’s the title of the consumer at the highest level of a food chain?” And RankBrain’s usage of AI means it works differently than the other technologies in the search engine.
“The other signals, they’re all based on discoveries and insights that people in information retrieval have had, but there’s no learning,” Corrado said.
Keeping an edge in search is critical to Google, and making its systems smarter and better able to deal with ambiguous queries is one of the ways it can keep a grip on time-starved users, who are now mostly searching using their mobile devices. “If you say Google people think of search,” Corrado said.
RankBrain is one of the “hundreds” of signals that go into an algorithm that determines what results appear on a Google search page and where they are ranked, Corrado said. In the few months it has been deployed, RankBrain has become the third-most important signal contributing to the result of a search query, he said.
“I was surprised,” Corrado said. “I would describe this as having gone better than we would have expected.”
The addition of RankBrain to search is part of a half-decade-long push by Google into AI, as the company seeks to embed the technology into every aspect of its business. “Machine learning is a core transformative way by which we are rethinking everything we are doing,” said Google’s Chief Executive Officer Sundar Pichai on the company’s earnings call last week.
Smarter Than Your Average Engineer
So far, RankBrain is living up to its AI hype. Google search engineers, who spend their days crafting the algorithms that underpin the search software, were asked to eyeball some pages and guess which they thought Google’s search engine technology would rank on top. While the humans guessed correctly 70 percent of the time, RankBrain had an 80 percent success rate.
Typical Google users agree. In experiments, the company found that turning off this feature “would be as damaging to users as forgetting to serve half the pages on Wikipedia,” Corrado said.
Getting here wasn’t easy. The rollout of RankBrain represents a yearlong effort by a team that started with about five Google engineers, including search specialist Yonghui Wu, and deep-learning expert Thomas Strohmann. It took a long time to make sure the system was ranking things correctly.
The effort expanded to dozens of people after Amit Singhal, the company’s senior vice president of search, gave the green light for it to be rolled out across all of Google search in early 2015.
“It’s very carefully monitored,” Corrado said, nothing that Google periodically updates the system by feeding it a load of new data to help it better reason with new concepts.
Google’s decision to deploy AI into search shows that companies are starting to entrust their most valuable businesses to systems controlled in part by machine intelligence. Facebook Inc. uses AI techniques to filter the newsfeed that comprises the personalized homepage of the social network and Microsoft Corp. is using artificial intelligence to increase the capabilities of its Bing search engine. Microsoft declined to be more specific about whether it’s using a similar approach to Google.
“Search is the cornerstone of Google,” Corrado said. “Machine learning isn’t just a magic syrup that you pour onto a problem and it makes it better. It took a lot of thought and care in order to build something that we really thought was worth doing.”