Microsoft wants to use machine learning to improve battery life

Over the last few years as our portable devices get more and more powerful, battery life has started to take a bit of a hit and unfortunately, nobody has quite figured out how to make bigger, longer-lasting batteries for super thin and light products yet. As a result, companies have been looking in to alternatives, such as software optimizations to conserve as much power as possible.

One of these companies is Microsoft, which wants to offer a broader range of power options to laptop and tablet users. Rather than waiting on new technology to develop, Microsoft wants to squeeze the most out of what we have right now.

Microsoft explained its new thinking in a recent blog post, noting that it wants to come up with a new system that uses smarter software, along with multiple current battery technologies, in order to determine which ones are used and when. This way, different processing loads could be handled by a different battery. Looking at an Excel spreadsheet would use a different battery to a heavier load, like HD streaming or playback.

“Rather than waiting for the perfect battery, we’re using all the technology available right now,” said Ranveer Chandra, a principal researcher at Microsoft Research.

“The software-defined battery system takes a different approach. It combines several different kinds of batteries, all of which are optimized for different tasks, into the same computer. Then, it works with the operating system to figure out whether the user is, say, looking at Word documents or editing video footage, and applies the most efficient battery for that task”, the blog post reads.

This software system would make use of machine learning, allowing the computer to learn a user's habits and adapt accordingly. With these techniques, Microsoft could theoretically squeeze the most out of battery technology we already have, rather than waiting for better technology to present itself.

Discuss on our Facebook page, HERE.

KitGuru Says: Battery life conservation has been a big focus point for device makers recently, particularly since battery technology itself has become quite stagnant, meaning more and more optimizations need to be made on the software side to make the most of it. Microsoft's machine learning approach could help improve things further, if it decides to implement it with Windows devices. 

Become a Patron!