Android Police

Articles Tagged:

machine learning


Android O feature spotlight: Smart Text Selection automatically highlights the right words at a double tap

One of the new features in Android O that was recently announced at I/O is Smart Text Selection. Selecting the right text can be hard on a small screen, but Google's data shows that most of the time when we are selecting information it's something like an address or phone number. Now in Android O you can double tap text, and the selection automatically expands as needed. It even shows a convenient predictive shortcut so we can use it in another context, or save it. Of course, we can use the Assistant to do all those things if we remember, but it's good to have multiple ways of solving the same problem.

Read More

Suggested Sharing, Shared Libraries, and photo books in Google Photos utilize machine learning to group photos together

Google Photos is something many people use every day; the automatic backup feature is so convenient, and the free unlimited storage is a major selling point. At I/O 2017, Google unveiled three new features: Suggested Sharing, Shared Libraries, and photo books. All of these use Photos' excellent machine learning technology to group faces together.

Read More

The robots are winning: Google's neural network OCR gives names to the nameless (streets)

Everyone uses Google Maps, but not often do we consider where, exactly, all that data comes from. When a new road goes in, or a bypass, or the name of a street changes, it isn't as if your local city reaches out to Google to make sure everything is up to date. Some of that used to come from user submissions in the Map Maker forum. Now those tools are being rolled into Maps itself, but that's not the only source of information. Google's fleets of Street View cars collect an insane number of images, and nestled in with them are pictures of businesses, street signs, and addresses, and Google's latest research blog post goes into some interesting details about all that potential data.

Read More

Google AutoDraw uses machine learning to make a designer out of anyone

The latest experiment to come out of Google Creative Lab is a web-based tool that uses machine learning to help you with all your design needs. If you need to fire off a quick design for a birthday card or an event poster, AutoDraw has you covered.

Using the simple MSPaint-style interface, all you have to do is draw a rudimentary outline of whatever you would like, and the suggestion tool will guess what you’re drawing and offer you far more professional looking alternatives. Click on a suggestion and it automatically replaces your inferior handiwork with something much prettier, sourced from various talented artists, designers, and illustrators.

Read More

Google introduces Perspective, a machine learning initiative to help police comments

I've said before that comments on internet posts are a double-edged sword; on the one hand, you can have fantastic conversations with them or you get a toxic cesspit with people spitting whatever drivel comes to mind. Google is hoping to combat the latter with its newest machine learning initiative called Perspective.

Read More

Google's A.I. Duet will respond to music you play

Everyone enjoys making music of some kind, from children hitting pots and pans together to an experienced musician performing to a sell-out crowd. I'm tone deaf and even I enjoy it. Today, Google announced it has a new experiment, A.I. Duet, which uses machine learning to respond to music that you play.

A.I. Duet was built by developer and musician Yotam Mann, based on the Magenta framework Google released last year. Using this, he's been able to have artificial intelligence learn how to play music based on hundreds or thousands of samples, through machine learning. While computers have been able to play music for a while, they would have to have been taught how to play.

Read More

Google's TensorFlow hits v1.0 with upgrades to speed, flexibility, and production-readiness

At the TensorFlow Developer Summit, Google announced that its second-generation machine learning platform has reached the v1.0 milestone. In the time since it went open-source, TensorFlow has helped many professionals in several fields make huge strides in things like language translation, early skin cancer detection, and diabetic blindness prevention.

Read More

Cortana learns how to keep you on track, even if you forget to ask

We've all forgotten to do things that we promised to do. Luckily, we have these fancy personal assistants to help us stay on course. Cortana, Microsoft's offering (and whose name is way cooler than the actual product), is learning via machine learning to remind you of things... without you even asking.

Read More

Google's researchers explain the challenges and calculation process behind the new Parking Difficulty feature in Maps

A few weeks ago, a new Parking Difficulty icon started showing up in Google Maps 9.44 beta in some cities in the United States, then Google officially announced it and specified where it's available. It also said that the estimate is "based on historical parking data," similar to how traffic and popular times and visit durations are calculated, but it didn't go into a lot of details. Now those details are further clarified in a post on the Google Research Blog.

The difficulties of calculating parking availability stem from the many, many factors that can influence the equation: time and day and weather and holidays/events, permit or illegal parking in park-meter areas, vacant spots with paid park-meters from cars that left early, parking lots with multiple levels and different structures, and so on. Even with these issues accounted for, there's the false data coming from private/gated parking spots, taxis dropping users who look like they immediately found a parking and reached their destination, and bus stops showing up in the data sets as popular parking spots.

Read More

Photos on Google+ receive the machine learning treatment to help save on bandwidth usage

Machine learning is such a buzzword, or words, these days. Perhaps one of the biggest advocates of this technology, the almighty Google has been applying it to many of its services over the past few months. Now photographs on Google+ are getting some love with the addition of RAISR, which allows for low-resolution images to receive massive bumps in quality while saving on bandwidth.

Read More