谷歌DeepMind近日在人工智能领域再次迈出重要一步，推出了一款名为SIMA（可扩展、可指导、多世界智能体）的新型人工智能Agent，其设计目标是通过训练学习游戏技能，以更接近人类玩家的方式进行游戏。SIMA的研发标志着AI在游戏领域的智能表现达到了一个新的水平。
据《The Verge》报道，SIMA智能体目前正处于研究阶段，其独特之处在于能够通过训练适应不同类型的游戏，而不只是单纯地执行预设任务。在《Valheim：英灵神殿》、《无人深空》和《模拟山羊》等游戏中，SIMA已经展现出了学习和适应的能力，这表明它能够在复杂的游戏环境中逐步提升技能。
谷歌DeepMind的最终目标是让SIMA能够掌握任何类型的游戏，包括那些没有明确线性路径和终点的开放世界游戏。这一创新意味着AI智能体将不再局限于特定的游戏规则，而是能够在不断变化的环境中自我学习和进步，更接近人类玩家的游戏体验和策略形成。
SIMA的出现不仅对游戏行业产生深远影响，也为人工智能在其他领域，如决策制定、问题解决等方面的应用提供了新的可能。随着研究的深入，我们期待SIMA在未来能够带来更多的惊喜和突破。
英语如下：
**News Title:** “Google DeepMind Unveils SIMA: A New AI Breakthrough That Learns to Play Games Like Humans”
**Keywords:** Google DeepMind, SIMA Agent, Game Training
**News Content:**
Title: Google DeepMind Launches SIMA Agent for AI That Learns to Play Games More Human-like
Google DeepMind has taken another significant step in the realm of artificial intelligence with the introduction of a novel AI Agent called SIMA (Scalable, Instructable, Multi-World Agent). Designed to learn game skills in a manner resembling human players, SIMA signifies a new milestone in AI performance within the gaming industry.
According to a report by _The Verge_, the SIMA Agent is currently in the research phase. Its uniqueness lies in its ability to adapt to different game types through training, rather than merely executing pre-programmed tasks. Demonstrating its learning and adaptability, SIMA has shown prowess in games like _Valheim:英灵神殿_, _No Man’s Sky_, and _ Goat Simulator_, indicating its capacity to enhance its skills in complex gaming environments.
Google DeepMind’s ultimate objective is for SIMA to master any genre of game, including open-world games with no clear linear paths or endpoints. This innovation signifies that AI agents will no longer be bound by specific game rules but can learn and progress in dynamic environments, resembling human players’ gaming experiences and strategy formation.
The emergence of SIMA is set to have a profound impact on the gaming industry and opens up new possibilities for AI application in areas such as decision-making and problem-solving. As research progresses, we anticipate more surprises and breakthroughs from SIMA in the future.
【来源】https://www.theverge.com/2024/3/13/24099024/google-deepmind-ai-agent-sima-video-games
Views: 1