Google DeepMind’s SIMA 2 learns problem-solving in Goat Simulator 3
Google DeepMind has unveiled SIMA 2, an advanced gaming agent trained using its Gemini system, capable of navigating and solving challenges in diverse 3D environments. This development marks a significant advancement towards creating more generalized AI…
- Google DeepMind has unveiled SIMA 2, an advanced gaming agent trained using its Gemini system, capable of navigating and solving challenges in diverse 3D environments.
- This development marks a significant advancement towards creating more generalized AI systems.
- By utilizing Goat Simulator 3 as a training ground, DeepMind aims to enhance the agent’s adaptability and problem-solving skills in complex virtual spaces.