Search for: All records

Vision Transformers (ViTs) have evolved in the field of computer vision by transitioning traditional Convolutional Neural Networks (CNNs) into attention-based architectures. This architecture processes input images as sequences of patches. ViTs achieve enhanced performance in many tasks such as image classification and object detection due to their ability to capture global dependencies within input data. While their software implementations are widely adopted, deploying ViTs on hardware introduces several challenges. These include fault tolerance in the presence of hardware failures, real-time reliability, and high computational requirements. Permanent faults that are in processing elements, interconnections, or memory subsystems lead to incorrect computations and degrading system performance. This paper proposes a fault-tolerant hardware implementation of ViTs to overcome these challenges. This hardware implementation integrates real-time fault detection and recovery mechanisms. The architecture includes four primary units: patch embedding, encoder, decoder, and Multi Layer Perceptron (MLP) which are supported by fault-tolerant components such as lightweight recompute units, a centralized Built-In Self-Test (BIST), and a learning-based decision-making system using machine learning model 'decision tree'. These units are interconnected through a centralized global buffer for efficient data transfer, ensuring seamless operation even under fault conditions. 

Abstract The coloration mechanism of Oregon sunstone is a classic and controversial topic in mineralogy because of the unique coexistence of anisotropic (green-red) and isotropic (red) color zones within single feldspar crystals. After nearly 50 years of research, no models proposed to date have satisfactorily accounted for all observed optical phenomena. Here, we present high-resolution transmission electron microscopy analyses of samples prepared by focused ion beam extraction along specific crystal directions. In both the anisotropic and the isotropic color zones, we observed Cu nanoparticles (NPs) included within plagioclase but with different geometries. In the isotropic (red) zone, NPs were randomly distributed nano-spheres or nano-ellipsoids (8.7–12 nm in diameter) with an aspect ratio of 1–1.3. In contrast, in dichroic (green/red) zones, NPs were directionally aligned nano-rods (8.5–21 nm along the long axis) with an aspect ratio of ∼2.5. We applied localized surface plasmon resonance (LSPR) theory to simulate absorption spectra and developed a model to explain the observed optical properties. LA-ICP-MS and polarized UV-Vis spectroscopy were also performed to confirm our conclusions. This study systematically reveals the existence and optical influence of variably shaped metal-NP inclusions in feldspar crystals. Furthermore, it demonstrates the necessity of including LSPR in the canon of mineral coloration mechanisms. Cu-NP-bearing labradorite has been shown to exhibit third-order nonlinear optical properties, and approaches that incorporate NP shapes and sizes will assist in designing NP-embedded optical materials with tailored optical properties. 

Quench experiments on a unitary Bose gas around a broad Feshbach resonance have led to the discovery of universal dynamics. This universality manifests in the measured atomic momentum distributions, where, asymptotically, a quasiequilibrated metastable state is found in which both the momentum distribution and the timescales are determined by the particle density. In this Letter we present counterpart studies for the case of a very narrow Feshbach resonance of ${}^{133}\mathrm{Cs}$atoms with a width of 8.3 mG. In dramatic contrast to the behavior reported earlier, a rapid quench of an atomic condensate to unitarity is observed to ultimately lead to coherent oscillations involving dynamically produced condensed and noncondensed molecules and atoms. The same characteristic frequency, determined by the Feshbach coupling, is observed in all types of particles. To understand these quench dynamics and how these different particle species are created, we develop a beyond Hartree-Fock-Bogoliubov dynamical framework including a type of cross-correlation between atoms and molecules. This leads to quantitative consistency with the measured frequency. Our results, which can be applied to the general class of bosonic superfluids associated with narrow Feshbach resonances, establish a alternate paradigm for universal dynamics dominated by quantum many-body interactions.