DeepMind AI 'Learns' to Navigate London Tube

...

DeepMind has introduced a new "computer" that uses memories to answer questions and even navigate the London Underground.

The UK-based artificial intelligence firm, owned by Google parent company Alphabet, built a "differentiable neural computer" (DNC), which consults its own memories to acquire and dispense information—much like the human brain.

"We showed how neural networks and memory systems can be combined to make learning machines that can store knowledge quickly and reason about it flexibly," DeepMind researchers Alexander Graves and Greg Wayne wrote in a blog post. "These models … can learn from examples like neural networks, but they can also store complex data like computers."

The DNC, for example, can store a roster of sports players and read out each name individually. It can also process more complicated tasks like deducing ancestral relationships based on a family tree (video above).

The real test, however, is a graph—one of the most complex and general data structures—like the London Underground network. After feeding the computer a layout of the Tube, the machine was able to recall a memory of the map, then answer intricate questions: "Starting at Bond Street and taking the Central line in a direction for one stop, the Circle line in a direction for four stops, and the Jubilee line in a direction for two stops, at what stop do you end up?"

DNC can also plan routes given a query like, "How do you get from Moorgate to Piccadilly Circus?"

"Conventional neural networks in our comparisons either could not store the information, or they could not learn to reason in a way that would generalize to new examples," Graves and Wayne said.

The scientists found they could also train via reinforcement learning: let the computer produce actions but never show it the answer, like the children's game "hot or cold" (video below). In testing, DNC was able to store several subroutines—one per possible goal—in memory and execute one or another.

"The question of how human memory works is ancient and our understanding still developing," the researchers said. "We hope that DNCs provide both a new tool for computer science and a new metaphor for cognitive science and neuroscience: Here is a learning machine that, without prior programming, can organize information into connected facts and use those facts to solve problems."

Categories
GAMES
0 Comment

Leave a Reply

Captcha image


RELATED BY

  • 5300c769af79e

    Fort Conquer v 1.2.1 Hack MOD | Download Apps For Free

    Fort Conquer - remarkable and quite enjoyable arcade tactical style wall at the wall, where we need to beat the waves of many horrific monsters, so that their enemy to get a logo on the short game location.{Requires Android 2.
  • 5300c769af79e

    Android Distribution Updated for August 2016 - Marshmallow Now Over 15%

    Google updated the Android distribution numbers while we were headed home from NYC yesterday, and wouldn’t you know it, Marshmallow continues to climb.Up from last month’s 13.
  • 5300c769af79e

    Facebook Messenger Users, You Now 1,500 New Emoji to Play With

    An update for Facebook Messenger is completely revamping the emoji experience inside of the communication app, but it’s not just the addition of new emoji that has people excited.Inside of the update, FB is bringing 1,500 newly designed emoji to public, including gender-agnostic options and multi-colored emojis.
  • 5300c769af79e

    2016 Insider Threat Report

    Download If you feel like you're not sure how to navigate the ever-changing landscape of insider threats, download the 2016 Insider Threat Report to learn how over 500 information security experts are rising to the challenge.This annual report reveals:- The most popular approaches to threat analysis and monitoring user behavior- The most frequent barriers to insider threat management- How often companies feel truly successfully at identifying insider attacks