Hideto Ogawa

Hideto Ogawa
Hitachi, Ltd. · Research & Development Group

PhD (Information Science)

About

24
Publications
2,146
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
108
Citations
Introduction
Hideto Ogawa currently works at the Research & Development Group, Hitachi, Ltd.. Hideto does research in Information Science, Theory of Computation and Software Engineering. Their most recent publication is 'Policing functions for machine learning systems'.

Publications

Publications (24)
Article
Full-text available
Significant effort is being put into developing industrial applications for artificial intelligence (AI), especially those using machine learning (ML) techniques. Despite the intensive support for building ML applications, there are still challenges when it comes to evaluating, assuring, and improving the quality or dependability. The difficulty st...
Conference Paper
Great efforts are currently underway to develop industrial applications for artificial intelligence (AI), especially those using machine learning (ML) techniques. Despite the intensive support for building ML applications, there are still challenges when it comes to evaluating, assuring, and improving the quality or dependability. The difficulty st...
Preprint
As for software development by machine learning, a trained model is evaluated by using part of an existing dataset as test data. However, if data with characteristics that differ from the existing data is input, the model does not always behave as expected. Accordingly, to confirm the behavior of the model more strictly, it is necessary to create d...
Article
As one type of machine-learning model, a “decision-tree ensemble model” (DTEM) is represented by a set of decision trees. A DTEM is mainly known to be valid for structured data; however, like other machine-learning models, it is difficult to train so that it returns the correct output value (called “prediction value”) for any input value (called “a...
Preprint
Full-text available
As one type of machine-learning model, a "decision-tree ensemble model" (DTEM) is represented by a set of decision trees. A DTEM is mainly known to be valid for structured data; however, like other machine-learning models, it is difficult to train so that it returns the correct output value for any input value. Accordingly, when a DTEM is used in r...
Conference Paper
During operation of a system including a deep neural network (DNN), new input values to the DNN that were not included in the initial dataset retained during development may be given. In such a case, the DNN may be additionally trained with the new input values; however, that additional training may reduce the accuracy of the DNN in regard to the i...
Article
In recent years, a number of methods for verifying DNNs have been developed. Because the approaches of the methods differ and have their own limitations, we think that a number of verification methods should be applied to a developed DNN. To apply a number of methods to the DNN, it is necessary to translate either the implementation of the DNN or t...
Preprint
Full-text available
In recent years, a number of methods for verifying DNNs have been developed. Because the approaches of the methods differ and have their own limitations, we think that a number of verification methods should be applied to a developed DNN. To apply a number of methods to the DNN, it is necessary to translate either the implementation of the DNN or t...
Conference Paper
Machine learning (ML) systems typically involve complex decision making mechanisms while lack clear and concise specifications. Demonstrating the quality of ML systems therefore is a challenging task. We propose an approach combining formal methods and metamorphic testing for improving the quality of ML systems. In particular, our framework enables...
Conference Paper
Early computer system were dominated by those with centralized architecture. In the 1980's, since the scale of the system expanded and equipment was distributed, a system architecture suitable for distributed system was developed. Today, due to the spread of the Internet of Things (IoT) technology, it is required to flexibly arrange and collaborate...
Conference Paper
Software refactoring is an effective way for dealing with technical debt accumulated in a software system; however, regression risk prevents developers from refactoring their code. Herein, we propose an approach to check the equivalence of source code before and after refactoring using the AST-based model extracted from source code. In order to det...
Article
This paper presents a symbolic-execution-based approach and its implementation by POM/JLEC for checking the logical equivalence between two programs in the system replacement context. The primary contributions lie in the development of POM/JLEC, a fully automatic equivalence checker for Java enterprise systems. POM/JLEC consists of three main compo...
Article
We propose a method for checking the behavior preservation property of refactoring based on static analysis of source code. Our method detects difference of program structure using AST-based models that are extracted from the source code and transformed according to the refactoring patterns. We also built a tool named POM/EQ that checks C/C++ sourc...
Conference Paper
When a software failure is observed during testing or operation, developers traditionally execute the software program again for reproducing the failure to analyze the cause of the failure. However, failures are often hard to reproduce because they depend on factors that are hard to expressly control, such as concurrency and nondeterminism. This pa...
Article
Full-text available
A practical model-checking (MC) approach for fault analysis, that is one of the most cost-effective tasks in software development, is proposed. The proposed approach is based on a technique, named “Program-oriented Modeling” (POM) for extracting a model from source code. The framework of model extraction by POM provides configurable abstraction bas...
Conference Paper
Debugging is one of the most time-consuming tasks in software development. The application of a model-checking technique in debugging has strong potential to solve this problem. Here, lessons learned through our practical experiences with POM/MC are discussed. The aim of this proposed hypothesis-based method of debugging is not only to reproduce a...
Conference Paper
Software product line engineering has spread as a technique for promoting the efficient development of embedded products with many product line-ups. During the development of network switch products at Hitachi Metals, Ltd., the number of development man-months increased as the number of product line-ups increased. Therefore, we shifted our developm...
Conference Paper
In the development of soft real-time embedded systems, it is difficult to meet real-time constraints by just coordinating among scheduling priorities of tasks. We have developed a resource reservation framework which distributes demanded CPU resources to the tasks. It employs an aspect-oriented mechanism which weaves a resource management code into...
Conference Paper
We herein propose Program-oriented Modeling (POM), a novel approach to extract software models automatically from source code. The proposed approach is developed in order to satisfy the requirements of model abstraction for situations in which a model is used. The POM model extraction framework provides rule-based abstraction using the Meta-Object...
Conference Paper
In a system composed of multiple units, it is necessary for system testing to prepare different system configurations. The selection of system configurations requires product knowledge, experience and many man-hours. This paper proposes a method to select system configurations using feature analysis. In this method, the result of the feature analys...
Conference Paper
Model checking is a powerful technique for verifying the correctness of a system's specification. But even when the specification has beet? verified to he correct, there is still the question of whether the specification covers all the expected behaviors. One of the most important issues for verification is the sufficiency Of verification items. In...
Article
Full-text available
The early integration of test development into the system development process becomes more and more important. By doing so, design mistakes and implemen- tation faults can be detected in an early stage of the system design and imple- mentation process, i.e. before the newly developed system is shipped to the cus- tomer. This allows for reducing the...

Network

Cited By

Projects

Project (1)