NASA 与 IBM 发布开源月球基础模型，训练数据来自 17 年轨道器观测
NASA and IBM's open source lunar model turns 17 years of orbiter data into a foundation for lunar science
NASA 和 IBM 把月球轨道器 17 年的观测数据训成了一个开源模型，预测极地冰误差能降 22%，做月球研究的可以直接拿来用。
NASA 与 IBM 联合发布开源的 Lunar Foundation Model，用于月球科学研究。该模型基于近 200 万个图像瓦片数据包训练，其中大部分来自 Lunar Reconnaissance Orbiter 累计 17 年的观测数据。在预测月球极地冰沉积物时，其误差比参与对比的最强模型最多降低 22%。模型采用开源方式发布，供月球科学研究者使用。
NASA and IBM's open source lunar model turns 17 years of orbiter data into a foundation for lunar science
NASA and IBM have released the Lunar Foundation Model, one of the first open-source AI models for lunar science. Trained on nearly 2 million tile bundles, mostly from 17 years of Lunar Reconnaissance Orbiter data, it cuts the error in predicting polar ice deposits by up to 22 percent compared to the strongest model it was tested against. The article NASA and IBM's open source lunar model turns 17 years of orbiter data into a foundation for lunar science appeared first on The Decoder .
