Search

We study theoretically the feasibility of the semi-exclusive $^{12}$C($p,dp$)$X$ reaction for the observation of $\eta^\prime$ mesic nuclei using the transport model JAM. The semi-exclusive measurements of the ($p,d$) reaction with protons from $\eta^\prime$ non-mesonic two-body absorption ($\eta^\prime NN \to NN$) are found to be critically important for the observation of the $\eta^\prime$ bound states. The Green's function method is used to calculate the momentum spectrum of forward going deuterons corresponding to the excitation energy spectrum of the $\eta^\prime \otimes {}^{11}$C system in the semi-exclusive measurement. The semi-exclusive measurements are considered to be important in general for the $\eta^\prime$ mesic nucleus formation., Comment: 7 pages, 2 figures, 1 table

The novel technique of mean range bunching has been developed and applied at the projectile fragment separator FRS at GSI in four experiments of the FAIR phase-0 experimental program. Using a variable degrader system at the final focal plane of the FRS, the ranges of the different nuclides can be aligned, allowing to efficiently implant a large number of different nuclides simultaneously in a gas-filled stopping cell or an implantation detector. Stopping and studying a cocktail beam overcomes the present limitations of stopped-beam experiments. The conceptual idea of mean range bunching is described and illustrated using simulations. In a single setting of the FRS, 37 different nuclides were stopped in the cryogenic stopping cell and were measured in a single setting broadband mass measurement with the multiple-reflection time-of-flight mass spectrometer of the FRS Ion Catcher.

Recent research has shown that training low-rank neural networks can effectively reduce the total number of trainable parameters without sacrificing predictive accuracy, resulting in end-to-end speedups. However, low-rank model training necessitates adjusting several additional factorization hyperparameters, such as the rank of the factorization at each layer. In this paper, we tackle this challenge by introducing Cuttlefish, an automated low-rank training approach that eliminates the need for tuning factorization hyperparameters. Cuttlefish leverages the observation that after a few epochs of full-rank training, the stable rank (i.e., an approximation of the true rank) of each layer stabilizes at a constant value. Cuttlefish switches from full-rank to low-rank training once the stable ranks of all layers have converged, setting the dimension of each factorization to its corresponding stable rank. Our results show that Cuttlefish generates models up to 5.6 times smaller than full-rank models, and attains up to a 1.2 times faster end-to-end training process while preserving comparable accuracy. Moreover, Cuttlefish outperforms state-of-the-art low-rank model training methods and other prominent baselines. The source code for our implementation can be found at: https://github.com/hwang595/Cuttlefish., Comment: Accepted for presentation at MLSys 2023

Direct mass measurements of neutron-deficient nuclides around the $N=50$ shell closure below $^{100}$Sn were performed at the FRS Ion Catcher (FRS-IC) at GSI, Germany. The nuclei were produced by projectile fragmentation of $^{124}$Xe, separated in the fragment separator FRS and delivered to the FRS-IC. The masses of 14 ground states and two isomers were measured with relative mass uncertainties down to $1\times 10^{-7}$ using the multiple-reflection time-of-flight mass spectrometer of the FRS-IC, including the first direct mass measurements of $^{98}$Cd and $^{97}$Rh. A new $Q_\mathrm{EC} = 5437\pm67$ keV was obtained for $^{98}$Cd, resulting in a summed Gamow-Teller (GT) strength for the five observed transitions ($0^+\longrightarrow1^+$) as $B(\text{GT})=2.94^{+0.32}_{-0.28}$. Investigation of this result in state-of-the-art shell model approaches sheds light into a better understanding of the GT transitions in even-even isotones at $N=50$. The excitation energy of the long-lived isomeric state in $^{94}$Rh was determined for the first time to be $293\pm 21$ keV. This, together with the shell model calculations, allows the level ordering in $^{94}$Rh to be understood.

In this paper, we present a new neural architectural block for the vision domain, named Mixing Regionally and Locally (MRL), developed with the aim of effectively and efficiently mixing the provided input features. We bifurcate the input feature mixing task as mixing at a regional and local scale. To achieve an efficient mix, we exploit the domain-wide receptive field provided by self-attention for regional-scale mixing and convolutional kernels restricted to local scale for local-scale mixing. More specifically, our proposed method mixes regional features associated with local features within a defined region, followed by a local-scale features mix augmented by regional features. Experiments show that this hybridization of self-attention and convolution brings improved capacity, generalization (right inductive bias), and efficiency. Under similar network settings, MRL outperforms or is at par with its counterparts in classification, object detection, and segmentation tasks. We also show that our MRL-based network architecture achieves state-of-the-art performance for H&E histology datasets. We achieved DICE of 0.843, 0.855, and 0.892 for Kumar, CoNSep, and CPM-17 datasets, respectively, while highlighting the versatility offered by the MRL framework by incorporating layers like group convolutions to improve dataset-specific generalization.

Modern theories of physics tell that the vacuum is not an empty space. Hidden in the vacuum is a structure of anti-quarks $\bar{q}$ and quarks $q$. The $\bar{q}$ and $q$ pair has the same quantum number as the vacuum and condensates in it since the strong interaction of the quantum chromodynamics (QCD) is too strong to leave it empty. The $\bar{q}q$ condensation breaks the chiral symmetry of the vacuum. The expectation value $<\bar{q}q>$ is an order parameter. For higher temperature or higher matter-density, $|<\bar{q}q>|$ decreases reflecting the restoration of the symmetry. In contrast to these clear-cut arguments, experimental evidence is so far limited. First of all, the $\bar{q}q$ is nothing but the vacuum itself. It is neither visible nor perceptible. In this article, we unravel this invisible existence by high precision measurement of pionic atoms, $\pi^-$-meson-nucleus bound systems. Using the $\pi^-$ as a probe, we demonstrate that $|<\bar{q}q>|$ is reduced in the nucleus at 58% of the normal nuclear density by a factor of 77 $\pm$ 2% compared with that in the vacuum. This reduction indicates that the chiral symmetry is partially restored due to the extremely high density of the nucleus. The present experimental result clearly exhibits the existence of the hidden structure, the chiral condensate, in the vacuum.

A novel method for simultaneous measurement of masses, Q-values, isomer excitation energies, half-lives and decay branching ratios of exotic nuclei has been demonstrated. The method includes first use of a stopping cell as an ion trap, combining containment of precursors and decay-recoils for variable durations in a cryogenic stopping cell (CSC), and afterwards the identification and counting of them by a multiple-reflection time-of-flight mass spectrometer (MR-TOF-MS). Feasibility has been established by recording the decay and growth of $^{216}$Po and $^{212}$Pb (alpha decay) and of $^{119m2}$Sb (t$_{1/2}$ = 850$\pm$90 ms) and $^{119g}$Sb (isomer transition), obtaining half-lives and branching ratios consistent with literature values. Hardly any non-nuclear-decay losses have been observed in the CSC for up to $\sim$10 seconds, which exhibits its extraordinary cleanliness. For $^{119}$Sb, this is the first direct measurement of the ground and second isomeric state masses, resolving the discrepancies in previous excitation energy data. These results pave the way for the measurement of branching ratios of exotic nuclei with multiple decay channels., Comment: 11 pages, 9 figures, 3 tables. Submitted to European Physics Journal A

Scaling the distributed deep learning to a massive GPU cluster level is challenging due to the instability of the large mini-batch training and the overhead of the gradient synchronization. We address the instability of the large mini-batch training with batch-size control and label smoothing. We address the overhead of the gradient synchronization with 2D-Torus all-reduce. Specifically, 2D-Torus all-reduce arranges GPUs in a logical 2D grid and performs a series of collective operation in different orientations. These two techniques are implemented with Neural Network Libraries (NNL). We have successfully trained ImageNet/ResNet-50 in 122 seconds without significant accuracy loss on ABCI cluster.

We are going to perform an inclusive spectroscopy experiment of eta' mesic nuclei with the 12C(p,d) reaction to study in-medium properties of the eta' meson. In nuclear medium, the eta' meson mass may be reduced due to partial restoration of chiral symmetry. In case of sufficiently large mass reduction and small absorption width of eta' at normal nuclear density, peak structures of eta' mesic states in 11C will be observed near the eta' emission threshold even in an inclusive spectrum. The experiment will be carried out at GSI with proton beam supplied by SIS using FRS as a spectrometer. The detail of the experiment is described., Comment: 4 pages, 2 figures, to appear in the proceedings of FB20 (20th International IUPAP Conference on Few-Body Problems in Physics), 20-25 August 2012, Fukuoka, Japan

The eta' meson mass may be reduced due to partial restoration of chiral symmetry. If this is the case, an eta'-nucleus system may form a nuclear bound state. We plan to carry out a missing-mass spectroscopy with the 12C(p,d) reaction at GSI-SIS. Peak structures corresponding to such a bound state may be observed even in an inclusive measurement, if the decay width is narrow enough., Comment: 4 pages, 2 figures, to appear in the proceedings of MESON2012 (12th International Workshop on Meson Production, Properties and Interaction), Krakow, Poland

A novel method is proposed to measure eta'(958) meson bound states in 11C nuclei by missing mass spectroscopy of the 12C(p,d) reaction near the eta' production threshold. It is shown that peak structures will be observed experimentally in an inclusive measurement in case that the in-medium eta' mass reduction is sufficiently large and that the decay width of eta' mesic states is narrow enough. Such a measurement will be feasible with the intense proton beam supplied by the SIS synchrotron at GSI combined with the good energy resolution of the fragment separator FRS., Comment: 12 pages, 6 figures, accepted for publication in Progress of Theoretical Physics