If you were curious how to predict your arrival time when using a traffic application, Google explains how it has improved the functionality of Google Maps with the help of Google Maps. Automatic learning from DeepMind.
Google Maps is one of the company's most widely used products, and its ability to predict traffic jams makes it indispensable for many customers.
According to Google, every day, more than 1 billion kilometers of roads are covered by the application. However, as explained in a blog post, its features have become more accurate thanks to the machine learning tools from DeepMind, the London-based AI lab owned by the company. mother of the giant Google, Alphabet Inc.
In the blog post, Google and DeepMind researchers explain how they retrieve data from a variety of sources and introduce it into machine learning models to predict traffic flows.
This data includes live traffic information, collected anonymously from Android devices, historical traffic data, information such as speed limits and antiquities from local authorities, and also factors such as the quality, size and direction of a given road.
All this information is entered into the neural networks designed by DeepMind, which select the patterns in the data and use them to predict future traffic, informs The Verge.
Google says its new models have improved the real-time ETA (estimated time on arrival) real-time accuracy of Google Maps by up to 50% in some cities.
Ã¢ â¬ ÅWe saw a drop of up to 50% in global traffic when quarantine measures began in early 2020, Ã¢ â¬ writes Johann Lau, Google Maps product manager.
Given this sudden change, we've updated our models to become more agile, automatically prioritizing historical traffic patterns from the last two to four weeks and archiving models at any time. anterior.Ã¢ â¬