Related papers: Machine Learning Tools for the IceCube-Gen2 Optica…

The IceCube Neutrino Observatory instruments roughly one cubic kilometer of deep, glacial ice below the geographic South Pole with 5160 optical sensors to register the Cherenkov light of passing relativistic, charged particles. Since its…

The difficult problem of relating the static structure of glassy liquids and their dynamics is a good target for Machine Learning, an approach which excels at finding complex patterns hidden in data. Indeed, this approach is currently a hot…

With construction halfway complete, IceCube is already the most sensitive neutrino telescope ever built. A rearrangement of the final holes of IceCube with increased spacing has been discussed recently to optimize the high energy…

Traditional analytical reflectance models, while compact and interpretable, lack the capacity to accurately represent physical measurements. Recent neural models, which closely fit input data, are less generalizable and often more expensive…

Neutrino telescopes allow powerful probes of high-energy astrophysics and particle physics. Their power is increased when they can isolate different event classes, e.g., by flavor, though that is not the only possibility. Here we focus on a…

Deep neural networks (DNNs) have substantial computational requirements, which greatly limit their performance in resource-constrained environments. Recently, there are increasing efforts on optical neural networks and optical computing…

This paper presents the custom implementation, optimization, and performance evaluation of convolutional neural networks on field programmable gate arrays, for the purposes of accelerating deep neural network inference on large,…

The IceCube Neutrino Observatory features both a kilometer-cubed detector between 1.45 and 2.45 km depth and an array of ice-filled tanks, called IceTop, located at the surface. The presence of both detectors at the same location allows for…

The discovery of ultra-high-energy neutrinos, with energies above 100 PeV, may soon be within reach of upcoming neutrino telescopes. We present a robust framework to compute the statistical significance of point-source discovery via the…

With the growing popularity of artificial intelligence used for scientific applications, the ability of attribute a result to a reasoning process from the network is in high demand for robust scientific generalizations to hold. In this work…

We demonstrate that a deep neural network can significantly improve optical microscopy, enhancing its spatial resolution over a large field-of-view and depth-of-field. After its training, the only input to this network is an image acquired…

Automated design methods for convolutional neural networks (CNNs) have recently been developed in order to increase the design productivity. We propose a neuroevolution method capable of evolving and optimizing CNNs with respect to the…

The soon-to-be-realized, global network of neutrino telescopes will allow new opportunities for collaboration between detectors. While each detector is distinct, they share the same underlying physical processes and detection principles.…

Convolutional Neural Networks (CNNs) are the state-of-the-art algorithms for the processing of images. However the configuration and training of these networks is a complex task requiring deep domain knowledge, experience and much trial and…

The IceCube Neutrino Observatory will be upgraded with about 700 additional optical sensor modules and new calibration devices. Particularly, improved calibration will enhance IceCube's physics capabilities both at low and high neutrino…

I discuss the feasibility of a conceptual space-based neutrino detector that utilizes the Ice Giants as Targets for Galactic Neutrinos. The purpose of this research stems from the concept of wanting to find a new method of observing the…

While the Standard Model has experienced great predictive success, the neutrino sector still holds opportunities for surprises. Numerous ongoing and planned experiments exist to probe neutrino properties at low energies. The IceCube…

A novel method, utilizing convolutional neural networks (CNNs), is proposed to reconstruct hyperspectral cubes from computed tomography imaging spectrometer (CTIS) images. Current reconstruction algorithms are usually subject to long…

IceCube is a one-gigaton instrument located at the geographic South Pole, designed to detect cosmic neutrinos, iden- tify the particle nature of dark matter, and study high-energy neutrinos themselves. Simulation of the IceCube detector and…

In recent years, deep neural network is introduced in recommender systems to solve the collaborative filtering problem, which has achieved immense success on computer vision, speech recognition and natural language processing. On one hand,…