arXiv subjects

Naonori Ueda

Publications and source records attributed to Naonori Ueda.

At least 19 recordsLinked to original sources

arXiv2025

Second-timescale Glints from Satellites and Space Debris Detected with Tomo-e Gozen

A search for second-timescale optical transients is one of the frontiers of time-domain astronomy. However, it has been pointed out that reflections of sunlight from Earth-orbiting objects can also produce second-timescale ``glints.'' We conducted wide-field observations at 2 frames per second using Tomo-e Gozen on the 1.05 m Kiso Schmidt telescope. We identified 1554 point-source glints that appeared in only one frame (0.5 sec). Their brightness ranges from 11 to 16 mag, with fainter glints being more numerous. These glints are likely caused by satellites and space debris in high-altitude orbits such as the geosynchronous Earth orbit or highly elliptical orbits. Many glints brighter than 14 mag are associated with known satellites or debris with large apogees ($>$ 30,000 km). In contrast, most fainter glints are not associated with cataloged objects and may be due to debris with sizes of 0.3--1 m. The event rate of second-timescale glints is estimated to be $4.7 \pm 0.2\ {\rm deg^{-2}\ hr^{-1}}$ (average) and $9.0 \pm 0.3\ {\rm deg^{-2}\ hr^{-1}}$ (near the equator) at 15.5 mag. Our results demonstrate that high-altitude debris can represent a significant foreground in searches for second-timescale optical transients. They also imply that deep surveys such as Rubin/LSST will detect many of these glints in single-exposure images.

astro-ph.EP↗

arXiv2025

Three-dimensional crustal deformation analysis using physics-informed deep learning

Earthquake-related phenomena such as seismic waves and crustal deformation impact broad regions, requiring large-scale modeling with careful treatment of artificial outer boundaries. Physics-informed neural networks (PINNs) have been applied to analyze wavefront propagation, acoustic and elastic waveform propagations, and crustal deformation in semi-infinite domains. In this study, we investigated the capability of PINNs for modeling earthquake crustal deformation in 3-D structures. To improve modeling accuracy, four neural networks were constructed to represent the displacement and stress fields in two subdomains divided by a fault surface and its extension. Forward simulations exhibited high accuracy for internal deformation but yielded errors for rigid motions, underscoring the inherent difficulty in constraining static deformation at an infinite distance. In the inversion analysis, fault slip distributions were estimated using surface observational data. Application to real data from the 2008 Iwate-Miyagi inland earthquake showed a fault slip consistent with previous studies, despite underestimation of the magnitude. This study demonstrates the capability of PINNs to analyze 3-D crustal deformation, thereby offering a flexible approach for large-scale earthquake modeling using real-world observations and crustal structures.

physics.geo-ph↗

arXiv2024

Neural Operators Meet Energy-based Theory: Operator Learning for Hamiltonian and Dissipative PDEs

The operator learning has received significant attention in recent years, with the aim of learning a mapping between function spaces. Prior works have proposed deep neural networks (DNNs) for learning such a mapping, enabling the learning of solution operators of partial differential equations (PDEs). However, these works still struggle to learn dynamics that obeys the laws of physics. This paper proposes Energy-consistent Neural Operators (ENOs), a general framework for learning solution operators of PDEs that follows the energy conservation or dissipation law from observed solution trajectories. We introduce a novel penalty function inspired by the energy-based theory of physics for training, in which the energy functional is modeled by another DNN, allowing one to bias the outputs of the DNN-based solution operators to ensure energetic consistency without explicit PDEs. Experiments on multiple physical systems show that ENO outperforms existing DNN models in predicting solutions from data, especially in super-resolution settings.

stat.ML↗

arXiv2023

Meta-learning of Physics-informed Neural Networks for Efficiently Solving Newly Given PDEs

We propose a neural network-based meta-learning method to efficiently solve partial differential equation (PDE) problems. The proposed method is designed to meta-learn how to solve a wide variety of PDE problems, and uses the knowledge for solving newly given PDE problems. We encode a PDE problem into a problem representation using neural networks, where governing equations are represented by coefficients of a polynomial function of partial derivatives, and boundary conditions are represented by a set of point-condition pairs. We use the problem representation as an input of a neural network for predicting solutions, which enables us to efficiently predict problem-specific solutions by the forwarding process of the neural network without updating model parameters. To train our model, we minimize the expected error when adapted to a PDE problem based on the physics-informed neural network framework, by which we can evaluate the error even when solutions are unknown. We demonstrate that our proposed method outperforms existing methods in predicting solutions of PDE problems.

stat.ML↗

arXiv2022

Excess risk analysis for epistemic uncertainty with application to variational inference

Bayesian deep learning plays an important role especially for its ability evaluating epistemic uncertainty (EU). Due to computational complexity issues, approximation methods such as variational inference (VI) have been used in practice to obtain posterior distributions and their generalization abilities have been analyzed extensively, for example, by PAC-Bayesian theory; however, little analysis exists on EU, although many numerical experiments have been conducted on it. In this study, we analyze the EU of supervised learning in approximate Bayesian inference by focusing on its excess risk. First, we theoretically show the novel relations between generalization error and the widely used EU measurements, such as the variance and mutual information of predictive distribution, and derive their convergence behaviors. Next, we clarify how the objective function of VI regularizes the EU. With this analysis, we propose a new objective function for VI that directly controls the prediction performance and the EU based on the PAC-Bayesian theory. Numerical experiments show that our algorithm significantly improves the EU evaluation over the existing VI methods.

stat.ML↗

arXiv2022

Deep-learning Real/Bogus classification for the Tomo-e Gozen transient survey

We present a deep neural network Real/Bogus classifier that improves classification performance in the Tomo-e Gozen transient survey by handling label errors in the training data. In the wide-field, high-frequency transient survey with Tomo-e Gozen, the performance of conventional convolutional neural network classifier is not sufficient as about $10^6$ bogus detections appear every night. In need of a better classifier, we have developed a new two-stage training method. In this training method, label errors in the training data are first detected by normal supervised learning classification, and then they are unlabeled and used for training of semi-supervised learning. For actual observed data, the classifier with this method achieves an area under the curve (AUC) of 0.9998 and a false positive rate (FPR) of 0.0002 at true positive rate (TPR) of 0.9. This training method saves relabeling effort by humans and works better on training data with a high fraction of label errors. By implementing the developed classifier in the Tomo-e Gozen pipeline, the number of transient candidates was reduced to $\sim$40 objects per night, which is $\sim$1/130 of the previous version, while maintaining the recovery rate of real transients. This enables more efficient selection of targets for follow-up observations.

astro-ph.IM↗

arXiv2021

Loss function based second-order Jensen inequality and its application to particle variational inference

Bayesian model averaging, obtained as the expectation of a likelihood function by a posterior distribution, has been widely used for prediction, evaluation of uncertainty, and model selection. Various approaches have been developed to efficiently capture the information in the posterior distribution; one such approach is the optimization of a set of models simultaneously with interaction to ensure the diversity of the individual models in the same way as ensemble learning. A representative approach is particle variational inference (PVI), which uses an ensemble of models as an empirical approximation for the posterior distribution. PVI iteratively updates each model with a repulsion force to ensure the diversity of the optimized models. However, despite its promising performance, a theoretical understanding of this repulsion and its association with the generalization ability remains unclear. In this paper, we tackle this problem in light of PAC-Bayesian analysis. First, we provide a new second-order Jensen inequality, which has the repulsion term based on the loss function. Thanks to the repulsion term, it is tighter than the standard Jensen inequality. Then, we derive a novel generalization error bound and show that it can be reduced by enhancing the diversity of models. Finally, we derive a new PVI that optimizes the generalization error bound directly. Numerical experiments demonstrate that the performance of the proposed PVI compares favorably with existing methods in the experiment.

stat.ML↗

arXiv2020

Photometric classification of HSC transients using machine learning

The advancement of technology has resulted in a rapid increase in supernova (SN) discoveries. The Subaru/Hyper Suprime-Cam (HSC) transient survey, conducted from fall 2016 through spring 2017, yielded 1824 SN candidates. This gave rise to the need for fast type classification for spectroscopic follow-up and prompted us to develop a machine learning algorithm using a deep neural network (DNN) with highway layers. This machine is trained by actual observed cadence and filter combinations such that we can directly input the observed data array into the machine without any interpretation. We tested our model with a dataset from the LSST classification challenge (Deep Drilling Field). Our classifier scores an area under the curve (AUC) of 0.996 for binary classification (SN Ia or non-SN Ia) and 95.3% accuracy for three-class classification (SN Ia, SN Ibc, or SN II). Application of our binary classification to HSC transient data yields an AUC score of 0.925. With two weeks of HSC data since the first detection, this classifier achieves 78.1% accuracy for binary classification, and the accuracy increases to 84.2% with the full dataset. This paper discusses the potential use of machine learning for SN type classification purposes.

astro-ph.IM↗

arXiv2020

Translation Between Waves, wave2wave

The understanding of sensor data has been greatly improved by advanced deep learning methods with big data. However, available sensor data in the real world are still limited, which is called the opportunistic sensor problem. This paper proposes a new variant of neural machine translation seq2seq to deal with continuous signal waves by introducing the window-based (inverse-) representation to adaptively represent partial shapes of waves and the iterative back-translation model for high-dimensional data. Experimental results are shown for two real-life data: earthquake and activity translation. The performance improvements of one-dimensional data was about 46% in test loss and that of high-dimensional data was about 1625% in perplexity with regard to the original seq2seq.

cs.LG↗

arXiv2020

Fully Neural Network based Model for General Temporal Point Processes

A temporal point process is a mathematical model for a time series of discrete events, which covers various applications. Recently, recurrent neural network (RNN) based models have been developed for point processes and have been found effective. RNN based models usually assume a specific functional form for the time course of the intensity function of a point process (e.g., exponentially decreasing or increasing with the time since the most recent event). However, such an assumption can restrict the expressive power of the model. We herein propose a novel RNN based model in which the time course of the intensity function is represented in a general manner. In our approach, we first model the integral of the intensity function using a feedforward neural network and then obtain the intensity function as its derivative. This approach enables us to both obtain a flexible model of the intensity function and exactly evaluate the log-likelihood function, which contains the integral of the intensity function, without any numerical approximations. Our model achieves competitive or superior performances compared to the previous state-of-the-art methods for both synthetic and real datasets.

cs.LG↗

arXiv2019

Anomaly Detection with Inexact Labels

We propose a supervised anomaly detection method for data with inexact anomaly labels, where each label, which is assigned to a set of instances, indicates that at least one instance in the set is anomalous. Although many anomaly detection methods have been proposed, they cannot handle inexact anomaly labels. To measure the performance with inexact anomaly labels, we define the inexact AUC, which is our extension of the area under the ROC curve (AUC) for inexact labels. The proposed method trains an anomaly score function so that the smooth approximation of the inexact AUC increases while anomaly scores for non-anomalous instances become low. We model the anomaly score function by a neural network-based unsupervised anomaly detection method, e.g., autoencoders. The proposed method performs well even when only a small number of inexact labels are available by incorporating an unsupervised anomaly detection mechanism with inexact AUC maximization. Using various datasets, we experimentally demonstrate that our proposed method improves the anomaly detection performance with inexact anomaly labels, and outperforms existing unsupervised and supervised anomaly detection and multiple instance learning methods.

stat.ML↗

arXiv2019

Deep Mixture Point Processes: Spatio-temporal Event Prediction with Rich Contextual Information

Predicting when and where events will occur in cities, like taxi pick-ups, crimes, and vehicle collisions, is a challenging and important problem with many applications in fields such as urban planning, transportation optimization and location-based marketing. Though many point processes have been proposed to model events in a continuous spatio-temporal space, none of them allow for the consideration of the rich contextual factors that affect event occurrence, such as weather, social activities, geographical characteristics, and traffic. In this paper, we propose \textsf{DMPP} (Deep Mixture Point Processes), a point process model for predicting spatio-temporal events with the use of rich contextual information; a key advance is its incorporation of the heterogeneous and high-dimensional context available in image and text data. Specifically, we design the intensity of our point process model as a mixture of kernels, where the mixture weights are modeled by a deep neural network. This formulation allows us to automatically learn the complex nonlinear effects of the contextual factors on event occurrence. At the same time, this formulation makes analytical integration over the intensity, which is required for point process estimation, tractable. We use real-world data sets from different domains to demonstrate that DMPP has better predictive performance than existing methods.

stat.ML↗

arXiv2019

The Hyper Suprime-Cam SSP Transient Survey in COSMOS: Overview

We present an overview of a deep transient survey of the COSMOS field with the Subaru Hyper Suprime-Cam (HSC). The survey was performed for the 1.77 deg$^2$ ultra-deep layer and 5.78 deg$^2$ deep layer in the Subaru Strategic Program over 6- and 4-month periods from 2016 to 2017, respectively. The ultra-deep layer shows a median depth per epoch of 26.4, 26.3, 26.0, 25.6, and 24.6 mag in $g$, $r$, $i$, $z$, and $y$ bands, respectively; the deep layer is $\sim0.6$ mag shallower. In total, 1,824 supernova candidates were identified. Based on light curve fitting and derived light curve shape parameter, we classified 433 objects as Type Ia supernovae (SNe); among these candidates, 129 objects have spectroscopic or COSMOS2015 photometric redshifts and 58 objects are located at $z > 1$. Our unique dataset doubles the number of Type Ia SNe at $z > 1$ and enables various time-domain analyses of Type II SNe, high redshift superluminous SNe, variable stars, and active galactic nuclei.

astro-ph.GA↗

arXiv2018

Unsupervised Object Matching for Relational Data

We propose an unsupervised object matching method for relational data, which finds matchings between objects in different relational datasets without correspondence information. For example, the proposed method matches documents in different languages in multi-lingual document-word networks without dictionaries nor alignment information. The proposed method assumes that each object has latent vectors, and the probability of neighbor objects is modeled by the inner-product of the latent vectors, where the neighbors are generated by short random walks over the relations. The latent vectors are estimated by maximizing the likelihood of the neighbors for each dataset. The estimated latent vectors contain hidden structural information of each object in the given relational dataset. Then, the proposed method linearly projects the latent vectors for all the datasets onto a common latent space shared across all datasets by matching the distributions while preserving the structural information. The projection matrix is estimated by minimizing the distance between the latent vector distributions with an orthogonality regularizer. To represent the distributions effectively, we use the kernel embedding of distributions that hold high-order moment information about a distribution as an element in a reproducing kernel Hilbert space, which enables us to calculate the distance between the distributions without density estimation. The structural information encoded in the latent vectors are preserved by using the orthogonality regularizer. We demonstrate the effectiveness of the proposed method with experiments using real-world multi-lingual document-word relational datasets and multiple user-item relational datasets.

stat.ML↗

arXiv2018

Finding Appropriate Traffic Regulations via Graph Convolutional Networks

Appropriate traffic regulations, e.g. planned road closure, are important in congested events. Crowd simulators have been used to find appropriate regulations by simulating multiple scenarios with different regulations. However, this approach requires multiple simulation runs, which are time-consuming. In this paper, we propose a method to learn a function that outputs regulation effects given the current traffic situation as inputs. If the function is learned using the training data of many simulation runs in advance, we can obtain an appropriate regulation efficiently by bypassing simulations for the current situation. We use the graph convolutional networks for modeling the function, which enable us to find regulations even for unseen areas. With the proposed method, we construct a graph for each area, where a node represents a road, and an edge represents the road connection. By running crowd simulations with various regulations on various areas, we generate traffic situations and regulation effects. The graph convolutional networks are trained to output the regulation effects given the graph with the traffic situation information as inputs. With experiments using real-world road networks and a crowd simulator, we demonstrate that the proposed method can find a road to close that reduces the average time needed to reach the destination.

physics.soc-ph↗

arXiv2018

Few-shot learning of neural networks from scratch by pseudo example optimization

In this paper, we propose a simple but effective method for training neural networks with a limited amount of training data. Our approach inherits the idea of knowledge distillation that transfers knowledge from a deep or wide reference model to a shallow or narrow target model. The proposed method employs this idea to mimic predictions of reference estimators that are more robust against overfitting than the network we want to train. Different from almost all the previous work for knowledge distillation that requires a large amount of labeled training data, the proposed method requires only a small amount of training data. Instead, we introduce pseudo training examples that are optimized as a part of model parameters. Experimental results for several benchmark datasets demonstrate that the proposed method outperformed all the other baselines, such as naive training of the target model and standard knowledge distillation.

stat.ML↗

arXiv2018

Partial AUC Maximization via Nonlinear Scoring Functions

We propose a method for maximizing a partial area under a receiver operating characteristic (ROC) curve (pAUC) for binary classification tasks. In binary classification tasks, accuracy is the most commonly used as a measure of classifier performance. In some applications such as anomaly detection and diagnostic testing, accuracy is not an appropriate measure since prior probabilties are often greatly biased. Although in such cases the pAUC has been utilized as a performance measure, few methods have been proposed for directly maximizing the pAUC. This optimization is achieved by using a scoring function. The conventional approach utilizes a linear function as the scoring function. In contrast we newly introduce nonlinear scoring functions for this purpose. Specifically, we present two types of nonlinear scoring functions based on generative models and deep neural networks. We show experimentally that nonlinear scoring fucntions improve the conventional methods through the application of a binary classification of real and bogus objects obtained with the Hyper Suprime-Cam on the Subaru telescope.

stat.ML↗

arXiv2017

Single-epoch supernova classification with deep convolutional neural networks

Supernovae Type-Ia (SNeIa) play a significant role in exploring the history of the expansion of the Universe, since they are the best-known standard candles with which we can accurately measure the distance to the objects. Finding large samples of SNeIa and investigating their detailed characteristics have become an important issue in cosmology and astronomy. Existing methods relied on a photometric approach that first measures the luminance of supernova candidates precisely and then fits the results to a parametric function of temporal changes in luminance. However, it inevitably requires multi-epoch observations and complex luminance measurements. In this work, we present a novel method for classifying SNeIa simply from single-epoch observation images without any complex measurements, by effectively integrating the state-of-the-art computer vision methodology into the standard photometric approach. Our method first builds a convolutional neural network for estimating the luminance of supernovae from telescope images, and then constructs another neural network for the classification, where the estimated luminance and observation dates are used as features for classification. Both of the neural networks are integrated into a single deep neural network to classify SNeIa directly from observation images. Experimental results show the effectiveness of the proposed method and reveal classification performance comparable to existing photometric methods with multi-epoch observations.

astro-ph.IM↗