2 min read

Earlier this month, AKQA, a global innovation agency, introduced a new outdoor sport called Speedgate, which is created by an AI system built by them. This AI system was trained on more than 400 sports including Rugby, Soccer, and football to form the rules and regulations for Speedgate.

In Speedgate, each team has six players consisting of forwards and defenders. The teams playing the game have to score goals by kicking through two consecutive gates. When a player kicks the ball through an X gate, the center gate will unlock the goal gate. After the center gate is unlocked, the team in possession can score by kicking the ball through the end gate in any direction.

Here’s a video showing how this game actually works:

Developers at AKQA trained a recurrent neural network and a deep convolutional generative adversarial network on over 400 sports. It uses NVIDIA Tesla GPUs for training the neural networks as well as for inferencing. Additionally, the neural network was also trained on 10,400 logos to generate the official Speedgate logo.

The model was able to generate over 1,000 different sport concepts. Though many of them were interesting, the team wanted the AI system to come up with a game that was in addition to being fun and easy to understand was also active and accessible. And, Speedgate checked all the boxes for them.

Kathryn Webb, AI Practice Lead at AKQA, said, “GPU technology helped us to condense training and generation phases down to a fraction of what they would’ve been. We would not have been able to achieve so many unique ML contributions in the project without that speed. It gave us more time to test, learn and adapt, and ultimately helped to produce the best final result.

Read more in detail, visit AKQA’s official website.

Read Next

OpenAI Five beats pro Dota 2 players; wins 2-1 against the gamers

Google announces Stadia, a cloud-based game streaming service, at GDC 2019

Microsoft announces Game stack with Xbox Live integration to Android and iOS