论文标题
部分可观测时空混沌系统的无模型预测
The INFN Experience in Supporting and Improving HEP Outreach
论文作者
论文摘要
储层计算是预测湍流的有力工具，其简单的架构具有处理大型系统的计算效率。然而，其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间，然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用，该系统模拟了湍流的若干特征。我们表明，使用径向基函数作为非线性投影器，即使只有部分观测并且不知道控制方程，也能稳健地捕捉复杂的系统非线性。最后，我们表明，当测量稀疏、不完整且带有噪声，甚至控制方程变得不准确时，我们的网络仍然可以产生相当准确的预测，从而为实际湍流系统的无模型预测铺平了道路。
INFN is recognized as an Italian excellence in science. Born in 1951, over time it created a world-wide network of activities spanning from high-energy physics at the most powerful accelerators, to the search for Dark Matter and rare events in deep underground laboratories, flanking the operations of four national laboratories in Italy. However, until a couple of decades ago, its role was not adequately appreciated by the Italian public at large. Since the beginning of the millennium INFN unfurled a strategy aimed not only to promote its image, but also to improve the transfer of knowledge acquired in its operation to different actors. In this paper we will deal with the improvement of outreach, presenting the strategies pursed, and some of the paths followed to this aim. Due to space limitations only a schematic view of a twenty-something years of work will be presented. In the conclusion we will report the results of an external evaluation of our efforts. In this paper we will not discuss the program of refresher courses for teachers, despite their relevance in our strategy to improve INFN participation in lifelong learning activities for the Italian society. Likewise, despite its centrality to improve INFN capability in the realm of outreach, we will not present the training program aimed to our personnel involved in science communication, and we will not touch the complex activity recently started to assess the impact of INFN communication efforts. Finally, we do not discuss the initiatives and the strategies pursued in 2020-2022 due to the COVID19 pandemia, despite the importance of outreach actions taken by the Institute.
