Back in 2014, Google acquired DeepMind Technologies Limited, a London-based company involved in research related to Artificial Intelligence. The company is famous for creating AlphaGo, the first computer program to beat a human professional player in a five-game match of Go.
Having acquired DeepMind for $500 million, it was evident that Google had big plans to incorporate the former’s AI research for the enhancement of the latter’s products. Fast forward to 2018, that research is helping in making Android Pie, the latest version of Google’s mobile operating system, even better.
Among the many features that Android 9.0 Pie has, two – Adaptive Brightness and Adaptive Battery – are aimed at improving battery life, an area where smartphones continue to be a disappointment. These features use Artificial Intelligence (AI) to automatically stop battery-hogging apps from constantly running in the background, as well as adjust display brightness based on ambient light. While the latter has been available on smartphones (through ambient light sensors) since forever, it’s now further enhanced by AI.
According to a report by Wired, Adaptive Brightness and Adaptive Battery are the result of the work done by DeepMind, in collaboration with Google.
Quoting Android engineer Ben Murdoch, the report says that the data collected from all beta versions and stable release of Android Pie show that the system is working.
“We have seen a reduction in what we call the variance. Most users are familiar with the day every now and then where your battery seems to drain much faster than you’re expecting, or it normally does. We call these unpredictable events ‘bad battery days’. We’re reining in those bad battery days,” commented Murdoch.
Detailing the basics of the Artificial Intelligence behind Android Pie’s battery saving features, Chris Gamble, who works in DeepMind’s Google products team, is quoted by the report as saying, “The model is a deep convolutional neural network. It’s got two layers and it uses timestamps of the app launches to predict when the app is going to be launched next.”
Watch: Vivo V11 Pro First Look
Once the AI classifies apps based on their launch times, they are placed into one of four buckets. Each bucket has restrictions in place, which prevent certain smartphone actions from happening.
While all of this sounds fun, the real tech working behind the scenes to make Adaptive Brightness and Adaptive Battery work is far more complicated. However, given how diverse each Android user and their usage habits are, perhaps only an AI is capable enough to predict behaviours to enhance battery life.