Reconstruction Techniques in IceCube using Convolutional and Generative Neural Networks
Reconstruction Techniques in IceCube using Convolutional and Generative Neural Networks
复制标题
IceCube 中使用卷积和生成神经网络的重建技术
DOI：
10.1051/epjconf/201920705005
复制
发表时间：
2019
影响因子：
--
通讯作者：
for the IceCube collaboration
中科院分区：
文献类型：
--
作者：
M. Huennefeld;for the IceCube collaboration
可靠和准确的重建方法对于像IceCube这样的高能物理实验的成功至关重要。基于机器学习的技术，特别是深度神经网络，可以为最大似然方法提供可行的替代方案。然而，大多数常见的神经网络架构都是为其他领域开发的，例如图像识别。虽然这些方法可以提高IceCube的重建性能，但定制技术仍有很大的潜力。在典型的物理用例中，数据中存在许多对称性，不变性和先验知识，这些都没有被当前的网络架构充分利用。提出了一种基于卷积神经网络的重建方法，与Ice-Cube中的标准重建方法相比，该方法可以显著提高重建精度，同时大大减少运行时间。此外，第一结果的基础上生成神经网络的未来发展进行了讨论。
Reliable and accurate reconstruction methods are vital to the success of high-energy physics experiments such as IceCube. Machine learning based techniques, in particular deep neural networks, can provide a viable alternative to maximum-likelihood methods. However, most common neural network architectures were developed for other domains such as image recogntion. While these methods can enhance the reconstruction performance in IceCube, there is much potential for tailored techniques. In the typical physics use-case, many symmetries, invariances and prior knowledge exist in the data, which are not fully exploited by current network architectures. Novel and specialized deep learning based reconstruction techniques are desired which can leverage the physics potential of experiments like IceCube.A reconstruction method using convolutional neural networks is presented which can significantly increase the reconstruction accuracy while greatly reducing the runtime in comparison to standard reconstruction methods in Ice- Cube. In addition, first results are discussed for future developments based on generative neural networks.