Google Deepmind Announced SIMA, A New AI Agent
On Wednesday, the Google Deepmind team introduced their new AI agent for 3D virtual settings- SIMA. Google Deepmind is the research team that has announced its new research project SIMA to the tech world.
Introduction to SIMA:
SIMA, short form for Scalable Instructable Multiworld Agent, is a co-op buddy that not only plays across a variety of 3D games but also listens, learns, and adapts to your instructions. SIMA is going to be your dream partner in the virtual world of games. This new AI agent is an easy learner.
SIMA has trained by watching humans playing different video games like Goat Simulator 3. It learns how humans interact with different characters in the games and easily adapt their actions and movements.
“We present new research on a Scalable Instructable Multiworld Agent (SIMA) that can follow natural-language instructions to carry out tasks in a variety of video game settings.”- The SIMA team shared in their blog post.
Also, read – Google Gemini will not Answer your Questions related to Elections 2024
“SIMA isn’t trained to win a game; it’s trained to run it and do what it’s told,” said Google DeepMind researcher and SIMA co-lead Tim Harley during a briefing with reporters.
SIMA: More than an AI agent
Google DeepMind trains a video game in collaboration with eight game studios. Google has a partnership with game developers so that they can train SEMA on different game environments, such as No Man’s Sky by Hello Games and Teardown by Tuxedo Labs.
Harley said-” the team chose games that were more focused on open play than narrative to help SIMA learn general gaming skills. If you’ve played or watched a playthrough of Goat Simulator, you know that doing random, spontaneous things is the point of the game, and Harley said it was this kind of spontaneity they hoped SIMA would learn.”
In their blog post, the SIMA search team has said that they have evaluated the current version of SIMA on around 600 basic skills, spanning navigation (e.g. “turn left”), object interaction (“climb the ladder”), and menu use (“open the map”). They said that SIMA is trained to perform simple tasks that can be completed within about 10 seconds.
The team also cleared that “Our AI agent doesn’t need access to a game’s source code, nor bespoke APIs. It requires just two inputs: the images on the screen, and simple, natural-language instructions provided by the user. SIMA uses keyboard and mouse outputs to control the games’ central character to carry out these instructions. This simple interface is what humans use, meaning SIMA can potentially interact with any virtual environment.”
