Search

This paper proposes a new approach using the stochastic projected gradient method and Malliavin calculus for optimal reinsurance and investment strategies. Unlike traditional methodologies, we aim to optimize static investment and reinsurance strategies by directly minimizing the ruin probability. Furthermore, we provide a convergence analysis of the stochastic projected gradient method for general constrained optimization problems whose objective function has H\"older continuous gradient. Numerical experiments show the effectiveness of our proposed method.

We numerically investigate the stress distribution in pastes after horizontal rotation by using an elasto-plastic model. Residual stress remains as a memory of rotation. The stress in the circumferential direction increases after the rotation, whereas that in the radial direction decreases. The residual stress is analytically related to the plastic deformation induced by the rotation. Based on the time evolution of plastic deformation, we theoretically describe the mechanism of the changes in the stress distribution.

In this study, we consider the problem of predicting task success for open-vocabulary manipulation by a manipulator, based on instruction sentences and egocentric images before and after manipulation. Conventional approaches, including multimodal large language models (MLLMs), often fail to appropriately understand detailed characteristics of objects and/or subtle changes in the position of objects. We propose Contrastive $\lambda$-Repformer, which predicts task success for table-top manipulation tasks by aligning images with instruction sentences. Our method integrates the following three key types of features into a multi-level aligned representation: features that preserve local image information; features aligned with natural language; and features structured through natural language. This allows the model to focus on important changes by looking at the differences in the representation between two images. We evaluate Contrastive $\lambda$-Repformer on a dataset based on a large-scale standard dataset, the RT-1 dataset, and on a physical robot platform. The results show that our approach outperformed existing approaches including MLLMs. Our best model achieved an improvement of 8.66 points in accuracy compared to the representative MLLM-based model., Comment: Accepted for presentation at CoRL2024

Analytic continuation (AC) from imaginary-time Green's function to spectral function is essential in the numerical analysis of dynamical properties in quantum many-body systems. However, this process faces a fundamental challenge: it is an ill-posed problem, leading to unstable spectra against the noise in the Green's function. This instability is further complicated in multi-orbital systems with hybridization between spin-orbitals, where off-diagonal Green's functions yield a spectral matrix with off-diagonal elements, necessitating the matrix's semi-positive definiteness to satisfy the causality. We propose an advanced AC method using sparse modeling for multi-orbital systems, which reduces the effect of noise and ensures the matrix's semi-positive definiteness. We demonstrate the effectiveness of this approach by contrasting it with the conventional sparse modeling method, focusing on handling Green's functions with off-diagonal elements, thereby demonstrating our proposed method's enhanced stability and precision., Comment: 8 pages, 8 figures

Correlations of the displacement field in a two-dimensional model colloidal liquid is studied numerically and analytically. By calculating the displacement correlations and the shear strain correlations from the numerical data of particle simulations, the displacement field is shown to have nontrivial correlations, even in liquids that are only slightly glassy with the area fraction as low as 0.5. It is suggested analytically and demonstrated numerically that the displacement correlations are more informative than the shear correlations: the former behaves logarithmically with regard to the spatial distance at shorter scales, while the corresponding information is missing from the shear correlations. The logarithmic behavior of the displacement correlations is interpreted as manifesting a long-lived aspect of the cage effect., Comment: 24 pages, 15 figures; to be published in Physical Review E

We numerically analyze the steady flow of two-dimensional granular materials in a rotating drum using the discrete element method and a continuum model. The velocity fields obtained from both methods are quantitatively consistent. Two distinct regions exist in the granular flow: the surface flow layer and the static flow regime associated with the rigid rotation near the bottom. The thickness of the surface flow layer increases with the drum diameter and shows a slight dependence on the angular velocity of the rotating drum. Scaling laws for the velocity profile and the surface layer thickness are derived analytically based on the continuum model. The validity of the scaling laws is confirmed numerically.

The transparent formulation of explanation methods is essential for elucidating the predictions of neural networks, which are typically black-box models. Layer-wise Relevance Propagation (LRP) is a well-established method that transparently traces the flow of a model's prediction backward through its architecture by backpropagating relevance scores. However, the conventional LRP does not fully consider the existence of skip connections, and thus its application to the widely used ResNet architecture has not been thoroughly explored. In this study, we extend LRP to ResNet models by introducing Relevance Splitting at points where the output from a skip connection converges with that from a residual block. Our formulation guarantees the conservation property throughout the process, thereby preserving the integrity of the generated explanations. To evaluate the effectiveness of our approach, we conduct experiments on ImageNet and the Caltech-UCSD Birds-200-2011 dataset. Our method achieves superior performance to that of baseline methods on standard evaluation metrics such as the Insertion-Deletion score while maintaining its conservation property. We will release our code for further research at https://5ei74r0.github.io/lrp-for-resnet.page/, Comment: Accepted for presentation at ECCV2024

Akali superoxides AO2 (A=Na, K, Rb, Cs), due to an open p shell of the oxygen ion O2^- with degenerate pi orbitals, have spin and orbital degrees of freedom. The complex magnetic, orbital, and structural phase transitions observed experimentally in this family of materials are only partially understood. Based on density functional theory, we derive a strong-coupling effective model for the isostructural compounds AO2 (A=K, Rb, Cs) from a two-orbital Hubbard model. We find that CsO2 has highly frustrated exchange interactions in the a-b plane, while the frustration is weaker for RbO2 and KO2. We solve the resulting Kugel-Khomskii model in the mean-field approximation. We show that CsO2 exhibits an antiferro-orbital (AFO) order with the ordering vector q=(1,0,0) and a stripe antiferromagnetic order with q=(1/2,0,0), which is consistent with recent neutron scattering experiments. We discuss the role of the pi-orbital degrees of freedom for the experimentally observed magnetic transitions and interpret the as-yet-unidentified T_s2=70K transition in CsO2 as an orbital ordering transition., Comment: 17 pages, 22 figures, 2 tables

We numerically study the flow of dense granular materials between parallel plates driven by an external force. The granular materials form a jammed solid-like state when the external force is below a critical force, while they flow like fluids above the critical force. The transition is characterized by the mass flux. The critical force depends on the average packing fraction and the distance between the plates. The scaling laws for the critical force and the mass flux are predicted theoretically based on a continuum model. They are numerically verified.

To understand the origin of dust in the circum-galactic medium (CGM), we develop a dust enrichment model. We describe each of the central galaxy and its CGM as a single zone, and consider the mass exchange between them through galactic inflows and outflows. We calculate the evolution of the gas, metal, and dust masses in the galaxy and the CGM. In the galaxy, we include stellar dust production and interstellar dust processing following our previous models. The dust in the galaxy is transported to the CGM via galactic outflows, and it is further processed by dust destruction (sputtering) in the CGM. We parameterize the time-scale or efficiency of each process and investigate the effect on the dust abundance in the CGM. We find that the resulting dust mass is sensitive to the dust destruction in the CGM, and the dust supply from galactic outflows, both of which directly regulate the dust abundance in the CGM. The inflow time-scale also affects the dust abundance in the CGM because it determines the gas mass evolution (thus, the star formation history) in the galaxy. The dust abundance in the CGM, however, is insensitive to stellar dust formation in the galaxy at later epochs because the dust production is dominated by dust growth in the interstellar medium. We also find that the resulting dust mass in the CGM is consistent with the value derived from a large sample of SDSS galaxies.

Wireless sensing is a promising technology for future wireless communication networks to realize various application services. Wireless local area network (WLAN)-based localization approaches using channel state information (CSI) have been investigated intensively. Further improvements in detection performance will depend on selecting appropriate feature information and determining the placements of distributed antenna elements. This paper presents a proposal of an enhanced device-free WLAN-based localization scheme with beam-tracing based antenna placement optimization and spatially localized regression, where beam-forming weights (BFWs) are used as feature information for training machine-learning (ML)-based models localized to partitioned areas. By this scheme, the antenna placement at the access point (AP) is determined by solving a combinational optimization problem with beam-tracing between AP and station (STA) without knowing the CSI. Additionally, we propose the use of localized regression to improve localization accuracy with low complexity, where classification and regression based ML models are used for coarse and precise estimations of the target position. We evaluate the proposed scheme effects on localization performance in an indoor environment. Experiment results demonstrate that the proposed antenna placement and localized regression scheme improve the localization accuracy while reducing the required complexity for both off-line training and on-line localization relative to other reference schemes., Comment: This work has been submitted to the IEEE for possible publication

The spin-dependent band structure of CoS$_2$ which is a candidate for a half-metallic ferromagnet was investigated by both spin- and angle-resolved photoemission spectroscopy and theoretical calculations, in order to reappraise the half-metallicity and electronic correlations. We determined the three-dimensional Fermi surface and the spin-dependent band structure. As a result, we found that a part of the minority spin bands is on the occupied side in the vicinity of the Fermi level, providing spectroscopic evidence that CoS$_2$ is not but very close to a half-metal. Band calculations using density functional theory with generalized gradient approximation showed a good agreement with the observed majority spin $e_g$ bands, while it could not explain the observed band width of the minority-spin eg bands. On the other hand, theoretical calculations using dynamical mean field theory could better reproduce the strong mass renormalization in the minority-spin $e_g$ bands. All those results strongly suggest the presence of anomalously enhanced spin-dependent electron correlation effects on the electronic structure in the vicinity of the half-metallic state. We also report the temperature dependence of the electronic structure across the Curie temperature and discuss the mechanism of the thermal demagnetization. Our discovery of the anomalously large spin-dependent electronic correlations not only demonstrates a key factor in understanding the electronic structure of half-metals but also provides a motivation to improve theoretical calculations on spin-polarized strongly correlated systems., Comment: 53 pages, 25 figures, 1 table

This study numerically investigated the friction of viscoelastic objects with grooves. A 3D viscoelastic block with grooves on a rigid substrate is slowly pushed from the lateral side under uniform pressure on the top surface. The local friction force at the interface between the block and the substrate obeys Amontons' law. Numerical results obtained using the finite element method reveal that the static friction coefficient decreases with increasing groove width and depth. The propagation of the precursor slip is observed before bulk sliding. Furthermore, bulk sliding occurs when the area of slow precursor slip reaches a critical value, which decreases with increasing groove size. A theoretical analysis based on a simplified model reveals that the static friction coefficient is related to the critical area of the precursor, which is determined by the instability of the precursor. A scaling law for the critical area is theoretically predicted, and it indicates that the decrease in the effective viscosity due to the formation of the grooves leads to a decrease in the static friction coefficient. The validity of the theoretical prediction is numerically confirmed., Comment: 25 pages, 15 figures; 2 ancillary files (Supplementary Videos) attached

Charged carriers in solids diffuse from hot to cold sides under temperature gradient to induce the thermoelectric voltage. Carrier filtering effect, which only passes either electrons or holes for the conduction process, is an efficient method to enhance such voltage, although it is challenging to experimentally realize it especially in conventional metals with weak energy dependence of the density of states near the Fermi level. Here we measure the in-plane and out-of-plane thermopower of the layered perovskite Sr$_2$RuO$_4$ single crystals above room temperature, and find that the out-of-plane thermopower is largely enhanced with increasing temperature, while the in-plane one seems to remain a temperature-independent constant value which is expected from the Heikes formula. The observed large out-of-plane thermopower may originate from the recently proposed intriguing hole filtering effect in the body-centered tetragonal system, in which the carrier hopping through the centered atom is essential. Thus, the present carrier filtering effect may be a universal property to be applicable in various materials belonging to such crystal system., Comment: 7 pages, 5 figures. Accepted for publication in Phys. Rev. Materials