Google AI program masters classic Atari video games

By remembering and learning, the DeepMind algorithm excels at retro games such as Breakout

In this screenshot of a Google DeepMind artificial intelligence program playing the Atari game Breakout, the program has learned to create a tunnel through the wall of blocks and send a ball through it to destroy more blocks, improving its score.

In this screenshot of a Google DeepMind artificial intelligence program playing the Atari game Breakout, the program has learned to create a tunnel through the wall of blocks and send a ball through it to destroy more blocks, improving its score.

Scientists at Google have produced an artificial intelligence program that can ace classic Atari video games.

The Deep Q-network (DQN) developed at London-based AI firm DeepMind, which was acquired by Google last year, can teach itself to play Atari 2600 video games using only the score and the pixel display as input information.

In a study published in the journal Nature, Demis Hassabis and other Google DeepMind collaborators said the program was able to surpass the performance of previous algorithms on most of the 49 games it tested on. It was also able to match the skill of a professional human games tester.

The program did not know the rules beforehand and was armed only with the motivation to maximize its score and the ability to learn from previous gaming sessions. It was able to excel at games including Video Pinball and Breakout, through the use of an algorithm that aims to imitate aspects of human thinking and learning.

The scientists used an approach known as reinforcement learning, which involves offering rewards as motivation for an AI system to learn. They combined that with a kind of artificial neural network, called a deep neural network, which makes use of various computational layers to represent increasingly abstract representations of data.

The team focused on a biologically inspired architecture known as a deep convolutional network, an approach similar to that taken by University of Maryland scientists who have been getting robots to teach themselves to cook by watching videos on YouTube.

The DQN algorithm did well on a variety of games, including side-scrolling shooter games, boxing matches and 3D car racing. It was also able to achieve more than 75 percent of the human score in 29 of the 49 games.

More significantly, however, it was able to learn strategy over many sessions. After 600 sessions playing Breakout, it learned the winning strategy of tunneling behind a wall of bricks that the player must destroy. It repeatedly sent the ball into the tunnel so it bounced around, destroying many bricks.

It's not the first time algorithms have been trained to play video games with minimal input, and DQN fared poorly on games such as Montezuma's Revenge, which requires a long-term planning strategy.

But the researchers said a single architecture has been shown to be able to learn and adapt when faced with various gaming challenges. They view it as another step in building effective, general-purpose AI programs.

"Taken together, our work illustrates the power of harnessing state-of-the-art machine learning techniques with biologically inspired mechanisms to create agents that are capable of learning to master a diverse array of challenging tasks," the authors wrote.

Tim Hornyak covers Japan and emerging technologies for The IDG News Service. Follow Tim on Twitter at @robotopia.

Join the Good Gear Guide newsletter!

Error: Please check your email address.

Tags popular scienceGooglegamesDeepMindinternetsearch engines

Our Back to Business guide highlights the best products for you to boost your productivity at home, on the road, at the office, or in the classroom.

Keep up with the latest tech news, reviews and previews by subscribing to the Good Gear Guide newsletter.

Tim Hornyak

IDG News Service
Show Comments

Most Popular Reviews

Latest News Articles


GGG Evaluation Team

Kathy Cassidy


First impression on unpacking the Q702 test unit was the solid feel and clean, minimalist styling.

Anthony Grifoni


For work use, Microsoft Word and Excel programs pre-installed on the device are adequate for preparing short documents.

Steph Mundell


The Fujitsu LifeBook UH574 allowed for great mobility without being obnoxiously heavy or clunky. Its twelve hours of battery life did not disappoint.

Andrew Mitsi


The screen was particularly good. It is bright and visible from most angles, however heat is an issue, particularly around the Windows button on the front, and on the back where the battery housing is located.

Simon Harriott


My first impression after unboxing the Q702 is that it is a nice looking unit. Styling is somewhat minimalist but very effective. The tablet part, once detached, has a nice weight, and no buttons or switches are located in awkward or intrusive positions.

Featured Content

Latest Jobs

Don’t have an account? Sign up here

Don't have an account? Sign up now

Forgot password?