The space and renewable energy tycoon has launched yet another pioneer non-profit project which can change the world of robotics today. Elon Musk developed a $1 billion-program "Robot Gym" to train artificial intelligence. He says he wanted the robots to be as versatile as humans when it comes to thinking.
Our reinforcement learning toolkit, OpenAI Gym, is now in public beta: https://t.co/8fMRLjSJQb.
— OpenAI (@open_ai) April 27, 2016
Tesla's CEO, together with Sam Altman, created the artificial intelligence non-profit research company. After a few months, OpenAI launched the gym to give developers a venue to train their AI machines according to a report by the International Business Times. Launched last week, the robot gym disseminates open source code to virtual playgrounds for artificial intelligence to play games, simulate exercises and perform specific tasks.
Greg Brockman and John Schulman of OpenAI were quoted, saying that, "Over time, we plan to greatly expand this collection of environments." They are open to contributions for new "environments" and exercises. And because it's an open source gym, and funded by a non-profit organization, anyone can use it for free.
According to OpenAI, the public Gym Beta can be accessed and downloaded from their official website. The gym was initially built as a tool to aid their own internal research, in line with the philanthropy record of Elon Musk, it is not surprising that they released it to the public free of charge. OpenAI said "We hope it will be just as useful for the broader community."
The robot gym serves as a toolkit for artificial intelligence to develop "reinforcement learning (RL) algorithms." Reinforcement learning is a part of a study on artificial intelligence which is focused on achieving a task instead of just following simple commands. Popular Science explains that it works by rewarding the AI once its algorithm is performed well. This system is used in video games.
With the robot gym, the AI will be put into different environments and will be instructed to perform a task. Popular video game Atari is one of the available environments in this OpenAI Project.
To measure the achievements of the robots, there will be a message board where top performers will be listed. However, it is not based on high scores, but on the versatility of the artificial intelligence. "It's not just about maximising score; it's about finding solutions which will generalise well."
The system is currently compatible with TensorFlow and Theano but OpenAI said they will be developing more versions to enable compatibility with other systems.
Apparently, this endeavor is widely supported by high-profile donors who include LinkedIn Corp Co-Founder Reid Hoffman and the Amazon Web Services.
© 2021 NatureWorldNews.com All rights reserved. Do not reproduce without permission.