Search

In the last two decades, sparse regularization methods such as the LASSO have been applied in various fields. Most of the regularization methods have one or more regularization parameters, and to select the value of the regularization parameter is essentially equal to select a model, thus we need to determine the regularization parameter adequately. Regarding the determination of the regularization parameter in the linear regression model, we often apply the information criteria like the AIC and BIC, however, it has been pointed out that these criteria are sensitive to outliers and tend not to perform well in high-dimensional settings. Outliers generally have a negative influence on not only estimation but also model selection, consequently, it is important to employ a selection method that is robust against outliers. In addition, when the number of explanatory variables is quite large, most conventional criteria are prone to select unnecessary explanatory variables. In this paper, we propose model evaluation criteria via the statistical divergence with excellence in robustness in both of parametric estimation and model selection. Furthermore, our proposed criteria simultaneously achieve the selection consistency with the robustness even in high-dimensional settings. We also report the results of some numerical examples to verify that the proposed criteria perform robust and consistent variable selection compared with the conventional selection methods., Comment: 34 pages, 3 figures

Error correction (EC) based on large language models is an emerging technology to enhance the performance of automatic speech recognition (ASR) systems. Generally, training data for EC are collected by automatically pairing a large set of ASR hypotheses (as sources) and their gold references (as targets). However, the quality of such pairs is not guaranteed, and we observed various types of noise which can make the EC models brittle, e.g. inducing overcorrection in out-of-domain (OOD) settings. In this work, we propose two fundamental criteria that EC training data should satisfy: namely, EC targets should (1) improve linguistic acceptability over sources and (2) be inferable from the available context (e.g. source phonemes). Through these criteria, we identify low-quality EC pairs and train the models not to make any correction in such cases, the process we refer to as conservative data filtering. In our experiments, we focus on Japanese ASR using a strong Conformer-CTC as the baseline and finetune Japanese LLMs for EC. Through our evaluation on a suite of 21 internal benchmarks, we demonstrate that our approach can significantly reduce overcorrection and improve both the accuracy and quality of ASR results in the challenging OOD settings., Comment: Accepted to EMNLP 2024 Industry Track

Human and/or asset tracking using an attached sensor units helps understand their activities. Most common indoor localization methods for human tracking technologies require expensive infrastructures, deployment and maintenance. To overcome this problem, environmental sounds have been used for infrastructure-free localization. While they achieve room-level classification, they suffer from two problems: low signal-to-noise-ratio (SNR) condition and non-uniqueness of sound over the coverage area. A microphone localization method was proposed using supervised spectral decomposition and spatial likelihood to solve these problems. The proposed method was evaluated with actual recordings in an experimental room with a size of 12 x 30 m. The results showed that the proposed method with supervised NMF was robust under low-SNR condition compared to a simple feature (mel frequency cepstrum coefficient: MFCC). Additionally, the proposed method could be easily integrated with prior distribution, which is available from other Bayesian localizations. The proposed method can be used to evaluate the spatial likelihood from environmental sounds., Comment: 6 pages, 6 figures, accepted to IEEE/SICE SII 2023

Properties of the nuclear equation of state (EoS) can be probed by measuring the dynamical properties of nucleus-nucleus collisions. In this study, we present the directed flow ($v_1$), elliptic flow ($v_2$) and stopping (VarXZ) measured in fixed target Sn + Sn collisions at 270 AMeV with the S$\pi$RIT Time Projection Chamber. We perform Bayesian analyses in which EoS parameters are varied simultaneously within the Improved Quantum Molecular Dynamics-Skyrme (ImQMD-Sky) transport code to obtain a multivariate correlated constraint. The varied parameters include symmetry energy, $S_0$, and slope of the symmetry energy, $L$, at saturation density, isoscalar effective mass, $m_{s}^*/m_{N}$, isovector effective mass, $m_{v}^{*}/m_{N}$ and the in-medium cross-section enhancement factor $\eta$. We find that the flow and VarXZ observables are sensitive to the splitting of proton and neutron effective masses and the in-medium cross-section. Comparisons of ImQMD-Sky predictions to the S$\pi$RIT data suggest a narrow range of preferred values for $m_{s}^*/m_{N}$, $m_{v}^{*}/m_{N}$ and $\eta$.

Transferring the knowledge of large language models (LLMs) is a promising technique to incorporate linguistic knowledge into end-to-end automatic speech recognition (ASR) systems. However, existing works only transfer a single representation of LLM (e.g. the last layer of pretrained BERT), while the representation of a text is inherently non-unique and can be obtained variously from different layers, contexts and models. In this work, we explore a wide range of techniques to obtain and transfer multiple representations of LLMs into a transducer-based ASR system. While being conceptually simple, we show that transferring multiple representations of LLMs can be an effective alternative to transferring only a single representation., Comment: Accepted to ICASSP 2024

We present a single-image numerical phase retrieval method for Zernike phase-contrast microscopy (ZPM) that addresses halo and shade-off artifacts, as well as the weak phase condition, without requiring hardware modifications. By employing a rigorous physical model of ZPM and a gradient descent algorithm for its inversion, we achieve quantitative ZPM imaging. Our approach is experimentally validated using biological cells and its quantitative nature is confirmed through comparisons with digital holography observations.

The domain shift in pathological segmentation is an important problem, where a network trained by a source domain (collected at a specific hospital) does not work well in the target domain (from different hospitals) due to the different image features. Due to the problems of class imbalance and different class prior of pathology, typical unsupervised domain adaptation methods do not work well by aligning the distribution of source domain and target domain. In this paper, we propose a cluster entropy for selecting an effective whole slide image (WSI) that is used for semi-supervised domain adaptation. This approach can measure how the image features of the WSI cover the entire distribution of the target domain by calculating the entropy of each cluster and can significantly improve the performance of domain adaptation. Our approach achieved competitive results against the prior arts on datasets collected from two hospitals., Comment: Accepted by IEEE ISBI'23

Estimating the layout of a room from a single-shot panoramic image is important in virtual/augmented reality and furniture layout simulation. This involves identifying three-dimensional (3D) geometry, such as the location of corners and boundaries, and performing 3D reconstruction. However, occlusion is a common issue that can negatively impact room layout estimation, and this has not been thoroughly studied to date. It is possible to obtain 3D shape information of rooms as drawings of buildings and coordinates of corners from image datasets, thus we propose providing both 2D panoramic and 3D information to a model to effectively deal with occlusion. However, simply feeding 3D information to a model is not sufficient to utilize the shape information for an occluded area. Therefore, we improve the model by introducing 3D Intersection over Union (IoU) loss to effectively use 3D information. In some cases, drawings are not available or the construction deviates from a drawing. Considering such practical cases, we propose a method for distilling knowledge from a model trained with both images and 3D information to a model that takes only images as input. The proposed model, which is called Shape-Net, achieves state-of-the-art (SOTA) performance on benchmark datasets. We also confirmed its effectiveness in dealing with occlusion through significantly improved accuracy on images with occlusion compared with existing models., Comment: Accepted by CVPR2023 workshop (CIVILS)

Semi-supervised domain adaptation is a technique to build a classifier for a target domain by modifying a classifier in another (source) domain using many unlabeled samples and a small number of labeled samples from the target domain. In this paper, we develop a semi-supervised domain adaptation method, which has robustness to class-imbalanced situations, which are common in medical image classification tasks. For robustness, we propose a weakly-supervised clustering pipeline to obtain high-purity clusters and utilize the clusters in representation learning for domain adaptation. The proposed method showed state-of-the-art performance in the experiment using severely class-imbalanced pathological image patches.