Denver news in your inbox

Catch up on the most important stories affecting your hometown with Axios Denver

Please enter a valid email.

Please enter a valid email.

Subscription failed
Thank you for subscribing!

Des Moines news in your inbox

Catch up on the most important stories affecting your hometown with Axios Des Moines

Please enter a valid email.

Please enter a valid email.

Subscription failed
Thank you for subscribing!

Minneapolis-St. Paul news in your inbox

Catch up on the most important stories affecting your hometown with Axios Minneapolis-St. Paul

Please enter a valid email.

Please enter a valid email.

Subscription failed
Thank you for subscribing!

Tampa-St. Petersburg news in your inbox

Catch up on the most important stories affecting your hometown with Axios Tampa-St. Petersburg

Please enter a valid email.

Please enter a valid email.

Subscription failed
Thank you for subscribing!

Please enter a valid email.

Please enter a valid email.

Subscription failed
Thank you for subscribing!

The AI system operating in maze environments with partial information. Credit: DeepMind

A machine learning system from Google's DeepMind can collect snapshots of a 3D scene taken from different angles and then predict what that environment will look like from a viewpoint it hasn't seen before, according to research published today in Science.

The big picture: Researchers want to create AIs that can build models of the world from data they've seen and then use those models to function in new environments. That capability could take an AI from the realm of learning about a space to understanding it — much the same way humans do — and is key to developing machines that can move autonomously through the world. (Think: driverless cars.)

The context: Computer vision — spurred by the availability of data and increased computing power — has rapidly advanced in the past six years. Many of the underlying algorithms largely learn via supervision: an algorithm is given a large dataset that is labeled with information (for example, about the object in a scene) and uses it to predict an output.

“Supervised learning has been super successful but it’s unsatisfying for two reasons. One, humans need to manually create the [training] datasets, which is expensive and they don’t capture everything. And two, it is not the way infants or higher mammals learn.”
— Ali Eslami, study author and researcher at DeepMind

Instead, researchers want to train machines to learn from unlabeled inputs that they process without any guidance from a human, and then to be able to apply or transfer what they learn to other new scenarios and tasks.

How it works: The system uses a pair of images of a virtual 3D scene taken from different angles to create a representation of the space. A separate “generation” network then predicts what the scene will look like from a different viewpoint it hasn’t seen before.

  • After training the generative query network (GQN) on millions of images, it could use one image to determine the identity, position and color of objects as well as shadows and other aspects of perspective, the authors wrote.
  • That ability to understand the scene's structure is the "most fascinating" part of the study, wrote the University of Maryland's Matthias Zwicker, who wasn't involved in the research.
  • The DeepMind researchers also tested the AI in a maze and reported the network can accurately predict a scene with only partial information.
  • A virtual robotic arm could also be controlled by the GQN to reach a colored object in a scene.

Yes, but: These are relatively simple virtual environments and "it remains unclear how close [the researchers'] approach could come to understanding complex, real-world environments," Zwicker writes.

Harvard's Sam Gershman told MIT Technology Review the GQN still solves only the narrow problem of predicting what a scene looks like from a different angle. According to the article:

"Gershman says it’s unclear whether DeepMind’s approach could be adapted to answer more complex questions or whether some fundamentally different approach might be required."

The challenges: Eslami says it took a couple of months to train the network. “We really were pushing the hardware available to us to its limits. We need a step up in hardware capabilities and the techniques to build these deep neural networks and train them.”

Go deeper: Read more about the various ways researchers are trying to design AI to work like the human brain.

Go deeper

Ina Fried, author of Login
3 mins ago - Technology

Apple sets September quarter sales record despite pandemic

Apple CEO Tim Cook, speaking at the Apple 12 launch event in October. Photo: Apple

Apple on Thursday reported quarterly sales and earnings that narrowly exceeded analysts estimates as the iPhone maker continued to see strong demand amid the COVID-19 pandemic.

What they's saying: The company said response to new products, including the iPhone 12 has been "tremendously positive" but did not give a specific forecast for the current quarter.

Supreme Court rejects second GOP effort to cut absentee ballot deadline in N.C.

Photo: Robert Alexander/Getty Images

The Supreme Court, for the second time in two days, rejected a GOP request to shorten the deadline mail-in ballots must be received by North Carolina officials to be counted.

The state of play: The state's deadline had been extended from 3 days to 9 days post-Election Day.

2 hours ago - Podcasts

The vaccine race turns toward nationalism

The coronavirus pandemic is worsening, both in the U.S. and abroad, with cases, hospitalizations and deaths all rising.

Axios Re:Cap digs into the state of global vaccine development — including why the U.S. and China seem to going at it alone — with medicinal chemist and biotech blogger Derek Lowe.