Rideshare giant and autonomous vehicle pioneer Uber has announced the creation of Uber AI Labs, a new division of Uber, based in San Francisco. The new division is a direct result of Uber’s acquisition of artificial intelligence (AI) research start-up Geometric Intelligence. The technology firm’s 15 employees will form the core of the AI Labs team, directed by Geometric’s founding CEO Gary Marcus, according to a press release issued by Uber.
Uber, AI and Self-Driving Cars
Uber has made great strides in the rollout and testing of autonomous vehicles recently, with fleets on the road in San Francisco, Phoenix and Pittsburgh collecting data to make self-driving cars even safer and more reliable. Geometric’s expertise and advancements in AI could make it possible for Uber autonomous vehicles to do more with less data.
An Uber self-driving car tours the streets of Pittsburgh. / Image credit: Uber That is the premise behind Geometric’s Xprop software, an advancement that permits computers to complete basic tasks with just one-quarter of the data most programs require. Still in development, the software is not yet as accurate as other programs using more data; but as it improves, it could help Uber self-driving cars reach the mainstream that much faster and reduce the amount of time required to “train” the vehicles in different cities.
Additionally, studies on deep generator networks are helping the experts at Geometric understand exactly what each neuron in a neural network is doing, further streamlining the learning process for AI. These advancements could also help correct driving habits that AI might mistakenly pick up from human drivers, such as over-correcting when a car starts to skid on icy roads.
Moving People and Things Faster
Uber has already disrupted an industry with the introduction of its ride-hailing service. This new acquisition and the formation of Uber AI Labs could put Uber at the front of the line in the autonomous vehicle race. The company is driven by the vision that “moving people and things in the physical world can be radically faster, safer and accessible to all,” said Chief Product Officer Jeff Holden in a statement.
