摘要
由机器人与机器学习每日新闻的新闻记者兼新闻编辑-研究人员详细介绍了人工智能的新数据。根据NewsRx编辑在巴塞罗那超级计算机G中心的新闻报道，研究表明，"在大规模地震之后，影响评估的一个关键任务是快速准确地估计受灾地区的地面震动。"我们的新闻编辑从巴塞罗那超级计算机中心的研究中获得了一句话：“为了满足实时约束，强度测量传统上都是用经验地震动模型来评估的，这种模型可以极大地限制估计值的准确性。作为替代方案，在此，我们介绍了在基于物理的模拟上训练的机器学习策略，这些模拟需要类似的评估时间。我们训练并验证了所提议的基于机器学习的估计器，用南加州地震中心开发的覆盖洛杉矶盆地的CyberShake现有最大数据集之一（<100M模拟地震图）对地面震动地图进行了训练和验证。我们的预测优于经验地震动模型，只要所考虑的事件与训练数据相一致。根据新闻编辑的说法，研究得出的结论是："相对于经验地震动模型，使用拟议的策略，我们不仅对合成地震，而且对五次真实的历史地震都显示出明显的误差减少。"
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Researchers detail new data in artific ial intelligence. According to news reporting out of the Barcelona Supercomputin g Center by NewsRx editors, research stated, “After large-magnitude earthquakes, a crucial task for impact assessment is to rapidly and accurately estimate the ground shaking in the affected region.” Our news editors obtained a quote from the research from Barcelona Supercomputin g Center: “To satisfy real-time constraints, intensity measures are traditionall y evaluated with empirical Ground Motion Models that can drastically limit the a ccuracy of the estimated values. As an alternative, here we present Machine Lear ning strategies trained on physics-based simulations that require similar evalua tion times. We trained and validated the proposed Machine Learning-based Estimat or for ground shaking maps with one of the largest existing datasets (<100M simulated seismograms) from CyberShake developed by the Southern California Earthquake Center covering the Los Angeles basin. For a well-tailored synthetic database, our predictions outperform empirical Ground Motion Models provided th at the events considered are compatible with the training data.” According to the news editors, the research concluded: “Using the proposed strat egy we show significant error reductions not only for synthetic, but also for fi ve real historical earthquakes, relative to empirical Ground Motion Models.”
NETL
NSTL