Positive and negative dependence are fundamental concepts that characterize the attractive and repulsive behavior of random subsets. Although some probabilistic models are known to exhibit positive or negative dependence, it is challenging to seamlessly bridge them with a practicable probabilistic model. In this study, we introduce a new family of distributions, named the discrete kernel point process (DKPP), which includes determinantal point processes and parts of Boltzmann machines. We also develop some computational methods for probabilistic operations and inference with DKPPs, such as calculating marginal and conditional probabilities and learning the parameters. Our numerical experiments demonstrate the controllability of positive and negative dependence and the effectiveness of the computational methods for DKPPs.

We propose new static word embeddings optimised for sentence semantic representation.We first extract word embeddings from a pretrained Sentence Transformer, and improve them with sentence-level principal component analysis, followed by either knowledge distillation or contrastive learning.During inference, we represent sentences by simply averaging word embeddings, which requires little computational cost.We evaluate models on both monolingual and cross-lingual tasks and show that our model substantially outperforms existing static models on sentence semantic tasks, and even surpasses a basic Sentence Transformer model (SimCSE) on a text embedding benchmark.Lastly, we perform a variety of analyses and show that our method successfully removes word embedding components that are not highly relevant to sentence semantics, and adjusts the vector norms based on the influence of words on sentence semantics.

The Bregman divergence, which is generated from a convex function, is commonly used as a pseudo-distance for comparing vectors or functions in continuous spaces. In contrast, defining an analog of the Bregman divergence for discrete spaces is nontrivial. Iyer & Bilmes (2012b) considered Bregman divergences on discrete domains using submodular functions as generating functions, the discrete analogs of convex functions. In this paper, we further generalize this framework to cases where the generating function is neither submodular nor supermodular, thus increasing the flexibility and representational capacity of the resulting divergence, which we term the difference-of-submodular Bregman divergence. Additionally, we introduce a learnable form of this divergence using permutation-invariant neural networks (NNs) and demonstrate through experiments that it effectively captures key structural properties in discrete data. As a result, the proposed method significantly improves the performance of existing methods on tasks such as clustering and set retrieval problems. This work addresses the challenge of defining meaningful divergences in discrete settings and provides a new tool for tasks requiring structure-preserving distance measures.

We propose a nonlinear probabilistic generative model of Koopman mode decomposition based on an unsupervised gaussian process. Existing data-driven methods for Koopman mode decomposition have focused on estimating the quantities specified by Koopman mode decomposition: eigenvalues, eigenfunctions, and modes. Our model enables the simultaneous estimation of these quantities and latent variables governed by an unknown dynamical system. Furthermore, we introduce an efficient strategy to estimate the parameters of our model by low-rank approximations of covariance matrices. Applying the proposed model to both synthetic data and a real-world epidemiological data set, we show that various analyses are available using the estimated parameters.

We develop a stochastic finite element method with ultra-large degrees of freedom that discretize probabilistic and physical spaces using unstructured second-order tetrahedral elements with double precision using a mixed-precision implicit iterative solver that scales to the full Fugaku system and enables fast Uncertainty Quantification (UQ). The developed solver designed to attain high performance on a variety of CPU/GPU-based supercomputers enabled solving 37 trillion degrees-of-freedom problem with 19.8% peak FP64 performance on full Fugaku (89.8 PFLOPS) with 87.7% weak scaling efficiency, corresponding to 224-fold speedup over the state of the art solver running on full Summit. This method, which has shown its effectiveness via solving huge (32-trillion degrees-of-freedom) practical problems, is expected to be a breakthrough in damage mitigation, and is expected to facilitate the scientific understanding of earthquake phenomena and have a ripple effect on other fields that similarly require UQ.

Dynamic mode decomposition (DMD) and its extensions are data-driven methods that have substantially contributed to our understanding of dynamical systems. However, because DMD and most of its extensions are deterministic, it is difficult to treat probabilistic representations of parameters and predictions. In this work, we propose a novel formulation of a Bayesian DMD model. Our Bayesian DMD model is consistent with the procedure of standard DMD, which is to first determine the subspace of observations, and then compute the modes on that subspace. Variational matrix factorization makes it possible to realize a fully-Bayesian scheme of DMD. Moreover, we derive a Bayesian DMD model for incomplete data, which demonstrates the advantage of probabilistic modeling. Finally, both of nonlinear simulated and real-world datasets are used to illustrate the potential of the proposed method.

The MPI library for the K computer introduced a memory saving protocol. However, the protocol still requires memory in proportion to the number of MPI processes and a memory shortage can occur when the number of processes reaches millions or tens of millions. In order to solve the problem, we propose the shared receive buffer method which is a new communication protocol using remote atomic operations. This method is easily implemented if an interconnect has remote memory access and remote atomic memory operation. We implemented shared receive buffer method on PRIMEHPC FX100 system and evaluated. The per process memory usage of the proposed method is about one tenth compared to that of existing method.

In this paper, we introduce a new and highly scalable interconnect called Tofu interconnect D that will be used in the post-K machine. This machine will officially be operational around 2021. The letter D represents high “density” node and “dynamic” packet slicing for “dual-rail” transfer. Herein we describe the design and the evaluation results of TofuD. Due to the high-density packaging, the optical link ratio of TofuD has decreased to 25% from the 66% optical link ratio of Tofu2. TofuD applies a new technique called dynamic packet slicing to reduce latency and to improve fault resilience. The evaluation results show that the one-way 8-byte Put latency is 0.49 μs. This is 31% lower than the latency of Tofu2. The injection rate per node is 38.1 GB/s which is approximately 83% of the injection rate of Tofu2. The link efficiency is as high as approximately 93%.

Hyperspectral Imager Suite (HISUI) is a next-generation Japanese optical sensor that is composed of a hyperspectral imager and a multispectral imager. Combination of hyperspectral and multispectral imagers contributes land remote sensing with high spectral and spatial resolution. Validation of the HISUI evaluation model, including environmental test such as vibration test and thermal-vacuum test has been carried out to meet the science requirements on the sensor under the space-borne conditions. Major performances of the sensors are obtained. Spectral response of each band of the hyperspectral sensor, distortions in a spectrometer in terms of smile and keystone properties, are investigated. The test results of the evaluation model show appropriate performance of the sensors.

Hyperspectral Imager Suite (HISUI) is a next-generation Japanese optical sensor that is composed of a hyperspectral imager and a multispectral imager, which will be launched on Advanced Land Observation Satellite 3 (ALOS-3). Combination of hyperspectral and multispectral imagers contributes land remote sensing with high spectral and spatial resolution. Calibration of the HISUI functional-equivalent model has been carried out to meet the science requirements on the sensor. A center wavelength and spectral response of each band of the hyperspectral sensor is obtained, from which distortions in a spectrometer in terms of smile and keystone properties are investigated. The results of the calibration activity for the functional-equivalent model show appropriate performance of the sensors.

The hyperspectral and the multispectral (hyper-multi spectral) mission is the Japanese next generation space-borne radiometer development project. This project is a heritage from ASTER launched in December 1999. The performance of the hyperspectral radiometer is 30m ground sampling distance, 30km swath width, 10nm and 12.5nm wavelength distance for VNIR and SWIR respectively, over 450@620nm and 300@2,100nm of the signal to noise (S/N) ratio. The performance of the multispectral radiometer is 5m ground sampling distance, 90km swath width, over 200 for all bands of S/N ratio. This project will be launched on ALOS-3 of JAXA in FY2014. The panchromatic sensor with stereo viewing will also be installed on ALOS-3.

OPUS is a nadir-looking, cross-track scanning ultraviolet spectrograph, which will be boarded on NASDA's GCOM-A1 satellite (one of the first generation satellites of Global Change Observation Mission), which is proposed to be launched in the year 2007. OPUS was formerly named Ozone Dynamics Ultraviolet Spectrometer (ODUS). Recently, we have come to place importance on observation of polluted atmospheric condition as well as continuous monitoring of ozone from NASA's Total Ozone Mapping Spectrometer (TOMS) series. Consequently, we have started to study about modification of instrument specifications. Currently, OPUS takes over the specification of ODUS as it is, but according to the result of the study, OPUS might be equipped with some new features. In this paper, the latest outcomes of the study are presented.

NASDA has been conducting an extended phase-A study on the Ozone and Pollution measuring Ultraviolet Spectrometer (OPUS), which is a nadir-looking, cross-track scanning ultraviolet spectrograph. OPUS was formerly named Ozone Dynamics Ultraviolet Spectrometer (ODUS). OPUS measures the backscattered solar radiation for the region between 306 and 420 nm with sampling step of 0.5 nm. It observes distributions of atmospheric ozone, sulfur dioxide (SO/sub 2/) and other parameters including aerosols. In this paper, (1) a feasibility of the retrieval of total column SO/sub 2/ is examined, and (2) a preliminary result on SO/sub 2/ measurement using airborne spectrograph is demonstrated.