{"id":372,"date":"2020-10-08T02:17:51","date_gmt":"2020-10-08T02:17:51","guid":{"rendered":"https:\/\/machine-learning.webcloning.com\/2020\/10\/08\/bada-bing-bada-boom-microsoft-turns-to-turing-nlg-nvidia-gpus-to-instantly-suggest-full-phrase-queries\/"},"modified":"2020-10-08T02:17:51","modified_gmt":"2020-10-08T02:17:51","slug":"bada-bing-bada-boom-microsoft-turns-to-turing-nlg-nvidia-gpus-to-instantly-suggest-full-phrase-queries","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2020\/10\/08\/bada-bing-bada-boom-microsoft-turns-to-turing-nlg-nvidia-gpus-to-instantly-suggest-full-phrase-queries\/","title":{"rendered":"Bada Bing Bada Boom: Microsoft Turns to Turing-NLG, NVIDIA GPUs to Instantly Suggest Full-Phrase Queries"},"content":{"rendered":"
\n

Hate hunting and pecking away at your keyboard every time you have a quick question? You\u2019ll love this.<\/p>\n

Microsoft\u2019s Bing search engine<\/a> has turned to Turing-NLG and NVIDIA GPUs to suggest full sentences for you as you type.<\/p>\n

Turing-NLG is a cutting-edge, large-scale unsupervised language model that has achieved strong performance on language modeling benchmarks.<\/p>\n

It\u2019s just the latest example of an AI technique called unsupervised learning<\/a>, which makes sense of vast quantities of data by extracting features and patterns without the need for humans to provide any pre-labeled data.<\/p>\n

Microsoft calls this Next Phrase Prediction, and it can feel like magic, making full-phrase suggestions in real time for long search queries.<\/p>\n

Turing-NLG is among several innovations \u2014 from model compression to state caching and hardware acceleration \u2014 that Bing has harnessed with Next Phrase Prediction.<\/p>\n

Over the summer, Microsoft worked with engineers at NVIDIA to optimize Turing-NLG to their needs, accelerating the model on NVIDIA GPUs to power the feature for users worldwide.<\/p>\n

A key part of this optimization was to run this massive AI model extremely fast to power real-time search experience. With a combination of hardware and model optimization Microsoft and NVIDIA achieved an average latency below 10 milliseconds.<\/p>\n

By contrast, it takes more than 100 milliseconds to blink your eye.<\/p>\n

Learn more about the next wave of AI innovations at Bing<\/a>.<\/p>\n

Before the introduction of Next Phrase Prediction, the approach for handling query suggestions for longer queries was limited to completing the current word being typed by the user.<\/p>\n

Now type in \u201cThe best way to replace,\u201d and you\u2019ll immediately see three suggestions for completing the phrase: wood, plastic and metal. Type in \u201chow can I replace a battery for,\u201d and you\u2019ll see \u201ciphone, samsung, ipad and kindle\u201d all suggested.<\/p>\n

\"\"<\/a><\/p>\n

With Next Phrase Prediction, Bing can now present users with full-phrase suggestions.<\/p>\n

The more characters you type, the closer Bing gets to what you probably want to ask.<\/p>\n

And because these suggestions are generated instantly, they\u2019re not limited to previously seen data or just the current word being typed.<\/p>\n

So, for some queries, Bing won\u2019t just save you a few keystrokes \u2014 but multiple words.<\/p>\n

As a result of this work, the coverage of autosuggestion completions increases considerably, Microsoft reports, improving the overall user experience \u201csignificantly.\u201d<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

http:\/\/feedproxy.google.com\/~r\/nvidiablog\/~3\/afZ7DLCSSGM\/<\/p>\n","protected":false},"author":0,"featured_media":373,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/372"}],"collection":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/comments?post=372"}],"version-history":[{"count":0,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/372\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media\/373"}],"wp:attachment":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media?parent=372"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/categories?post=372"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/tags?post=372"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}