Related papers: Evaluation of Xilinx Deep Learning Processing Unit…

We consider deep feedforward neural networks with rectified linear units from a signal processing perspective. In this view, such representations mark the transition from using a single (data-driven) linear representation to utilizing a…

Understanding the behavior of materials under irradiation is crucial for the design and safety of nuclear reactors, spacecraft, and other radiation environments. The threshold displacement energy (Ed) is a critical parameter for…

Currently, Deep learning and especially Convolutional Neural Networks (CNNs) have become a fundamental computational approach applied in a wide range of domains, including some safety-critical applications (e.g., automotive, robotics, and…

The precise measurement of neutrino oscillation parameters is one of the highest priorities in neutrino oscillation physics. To achieve the desired precision, it is necessary to reduce the systematic uncertainties related to neutrino energy…

Fault-tolerant deep learning accelerator is the basis for highly reliable deep learning processing and critical to deploy deep learning in safety-critical applications such as avionics and robotics. Since deep learning is known to be…

Deep convolutional neural networks (CNNs) have recently achieved great success for single image super-resolution (SISR) task due to their powerful feature representation capabilities. The most recent deep learning based SISR methods focus…

The Deep Underground Neutrino Experiment (DUNE) is a leading experiment in neutrino physics which is presently under construction. DUNE aims to measure the yet unknown parameters in the three flavour oscillation scenario which includes…

Recent studies have highlighted the sensitivity of core-collapse supernovae (CCSNe) models to electron-capture (EC) rates on neutron-rich nuclei near the N=50 closed-shell region. In this work, we perform a large suite of one-dimensional…

Training deep learning models is compute-intensive and there is an industry-wide trend towards hardware specialization to improve performance. To systematically benchmark deep learning platforms, we introduce ParaDnn, a parameterized…

Deep neural networks (DNNs) have achieved superior performance in various prediction tasks, but can be very vulnerable to adversarial examples or perturbations. Therefore, it is crucial to measure the sensitivity of DNNs to various forms of…

We explore the effects of non-standard neutrino interactions (NSI) and how they modify neutrino propagation in the Deep Underground Neutrino Experiment (DUNE). We find that NSI can significantly modify the data to be collected by the DUNE…

A deep convolutional neural network (CNN) is developed to study symmetry energy $E_{\rm sym}(\rho)$ effects by learning the mapping between the symmetry energy and the two-dimensional (transverse momentum and rapidity) distributions of…

The evaluation of new microprocessor designs is constrained by slow, cycle-accurate simulators that rely on unrepresentative benchmark traces. This paper introduces a novel deep learning framework for high-fidelity, ``in-the-wild''…

This work evaluates State-of-the-Art convolution algorithms for CPU-based CNN inference. Although most prior studies focus on GPUs or NPUs, CPU implementations remain comparatively under-optimized. Our first contribution is to provide fair…

Deep neural networks (DNN) consist of layers of neurons interconnected by synaptic weights. A high bit-precision in weights is generally required to guarantee high accuracy in many applications. Minimizing error accumulation between layers…

This paper presents an instruction-based coordination architecture for Field-Programmable Gate Array (FPGA)-based systems with multiple high-performance Processing Units (PUs) for accelerating Deep Neural Network (DNN) inference. This…

Deep neural networks (DNNs) have been shown to be vulnerable to adversarial attacks -- subtle, perceptually indistinguishable perturbations of inputs that change the response of the model. In the context of vision, we hypothesize that an…

Deep Neural Networks (DNNs) rely on inherent fluctuations in their internal parameters (weights and biases) to effectively navigate the complex optimization landscape and achieve robust performance. While these fluctuations are recognized…

This thesis describes an analysis developed for the MicroBooNE experiment to investigate an anomalous excess of electron-like events observed in the MiniBooNE detector. The hypothesis investigated here is that the MiniBooNE anomaly…

Research suggests that students often have difficulty transferring their learning from one context to another. We examine upper-level undergraduate and graduate students' facility with questions about the interference pattern in the…