Faculty Profiles - HONDA Aoi

Papers - HONDA Aoi

Division display1 - 20 of about 181／ All the affair displays >>

現代数学への誘い－人工知能入門数学で読み解く最新AI技術 Invited

本田あおい

現代数学 ( 現代数学社 ) 2024 ( 9 ) 2024.09

An experimental study of the relationship between t-norms and t-conorms in IEINNs Reviewed

Yoshihiro Fukushima, Katsushige Fujimoto, Aoi Honda

Proceedings The 21st International Conference on Modeling Decisions for Artificial Intelligence 30 - 41 2024.08

Inclusion-exclusion integral neural network using monotone regularization terms Reviewed

Hina Anai, Simon James, Aoi Honda

Proceedings The 21st International Conference on Modeling Decisions for Artificial Intelligence 42 - 53 2024.08

Hierarchical Feature Interpretation of Machine Learning Models: Introducing OHAP Approach Reviewed

Kenkou Yamashita, Katsushige Fujimoto, Aoi Honda

Proceedings The 21st International Conference on Modeling Decisions for Artificial Intelligence 99 - 107 2024.08

A new duality concept for AC and interaction operators through the IE-integral Reviewed International journal

Katsushige Fujimoto, Aoi Honda

Proceedings ot 20th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems 2024.07

包除積分（Inclusion-Exclusion Integral） Reviewed

本田あおい

知能と情報 ( 日本知能情報ファジィ学会 ) 36 ( 1 ) 45 - 45 2024.02

An Axiomatic Definition of Non-discrete Möbius transform Reviewed International journal

Fukuda R., Honda A., Okazaki Y.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 14986 LNAI 19 - 30 2024.01

A neural network based on the inclusion-exclusion integral and its application to data analysis Reviewed International journal

Honda A., Itabashi M., James S.

Information Sciences 648 119549 2023.11

A New Design Chart Method of Cylindrical Bearings Based on a Simplified Thermohydrodynamic Lubrication Model Reviewed

HATAKENAKA Kiyoshi, HONDA Aoi

JOURNAL OF JAPANESE SOCIETY OF TRIBOLOGISTS ( Japanese Society of Tribologists ) 68 ( 10 ) 706 - 709 2023.10

Program design and implementation of inclusion-exclusion integral neural network Reviewed International journal

Aoi Honda, Yoshihiro Fukushima

Proceedings The 20th International Conference on Modeling Decisions for Artificial Intelligence 98 - 109 2023.06

Constructive Set Function and Extraction of a k-dimensional Element Reviewed International journal

Fukuda R., Honda A., Okazaki Y.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 13890 LNCS 58 - 69 2023.05

Suitable Lp spaces for a k-additive set function Reviewed International journal

Fukuda R., Honda A., Okazaki Y.

Fuzzy Sets and Systems 457 20 - 31 2023.04

IE Integral Neural Network with Monotone Regularization Term

Anai Hina, Honda Aoi

Proceedings of the Fuzzy System Symposium ( Japan Society for Fuzzy Theory and Intelligent Informatics ) 39 ( 0 ) 804 - 809 2023.01

Authorship：Last author Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

The Inclusion-Exclusion integral neural network is a neural network model that uses fuzzy measures, which are non-additive measures, and Inclusion-Exclusion integrals. This makes it possible to measure the contribution of each explanatory variable using statistical indices such as Shapley values and has the advantage of preserving interpretability of the parameters after learning. However, to maintain interpretability, the monotonicity of fuzzy measures should be satisfied. In this study, we propose a regularization term that satisfies the monotonicity of the fuzzy measure for the parameters obtained by training an Inclusion-Exclusion integral neural network, and conduct experiments using regression data. Furthermore, we will verify under what conditions learning can be performed with high accuracy while preserving the monotonicity of the fuzzy measure, compared to the l1 regularization and l2 regularization that are currently widely used.

DOI： 10.14864/fss.39.0_804

CiNii Research

Faster Learning in Inclusive-Exclusive Integral Neural Networks

Fukushima Yoshihiro, Honda Aoi

Proceedings of the Fuzzy System Symposium ( Japan Society for Fuzzy Theory and Intelligent Informatics ) 39 ( 0 ) 810 - 814 2023.01

Authorship：Last author Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

Inclusion-exclusion integral neural network is an explainable network model using inclusion-exclusion integral defined by fuzzy measures and polynomial operations. Inclusion-exclusion integral neural networks increase the expressiveness of the network by specifying a large additivity, but the number of parameters increases exponentially, and the training time increases accordingly. This is one of the challenges. The objective of this research is to accelerate the learning of inclusion-exclusion integral neural networks by comparing the speed between CPU and GPU in terms of additivity and the number of data, and by applying acceleration methods in PyTorch.

DOI： 10.14864/fss.39.0_810

CiNii Research

Model interpretation of Inclusion-Exclusion integral neural network using SHAP Values and Owen Values

Yamashita Kenkou, Honda Aoi

Proceedings of the Fuzzy System Symposium ( Japan Society for Fuzzy Theory and Intelligent Informatics ) 39 ( 0 ) 798 - 803 2023.01

Authorship：Last author Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

Inclusion-Exclusion integral neural network is a network model that uses fuzzy measures, which are non-additive measures, and Inclusion-Exclusion integral, which are represented by polynomial operations. The model outputs of this network can be expressed in terms of Inclusion-Exclusion integral formula using the Möbius inversion formula, and the contribution of each feature to the prediction can be calculated using the Shapley values, which is one of the reward calculation methods used in the theory of cooperative games. This allows us to construct a network with interpretability for parameters after training. For this network, we interpret the model using the Shapley values and the Owen values, which is an extended version of the Shapley values when considering a particular coalition structure or division of the set of players. We will also examine model interpretation using SHAP values, where Shapley values are applied to machine learning to allow interpretation of the impact of features on prediction based on input-output relationships.

DOI： 10.14864/fss.39.0_798

CiNii Research

A New Design Method of Cylindrical Bearings Based on a Simplified Thermohydrodynamic Lubrication Model Reviewed

Hatakenaka Kiyoshi, Honda Aoi

Tribology Online ( Japanese Society of Tribologists ) 17 ( 3 ) 175 - 195 2022.09

Language：English Publishing type：Research paper (scientific journal)

This paper proposes a new THL bearing design method based on an improved version of a precedent method. The precedent method used a model expression prepared in advance to easily and quickly find a crucial predicted data of the maximum bearing temperature. Three expressions were prepared for three different values of journal eccentricity ratio, 0.25, 0.50, and 0.75. Which one of the three should be used was determined, based on the preliminarily obtained value of journal eccentricity ratio derived from the initially analysis and t–test, for the crucial temperature for any value of journal eccentricity ratio. Furthermore, the new expression explicitly includes the four well-known dimensionless numbers used in THL analysis, that is, Sommerfeld number, bearing width-to-dimeter ratio, Peclet number and Khonsari number. The model expressions for other bearing characteristic performances were obtained in the same way. Compared with some measurements in test rigs of cylindrical bearings, the new improved method is confirmed to predict reasonably accurate bearing performances of cylindrical bearings faster and more easily than the precedent method over a wide range of operating conditions.

DOI： 10.2474/trol.17.175

Scopus

CiNii Research

Other Link： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85140405454&origin=inward

Representation and Interpretability of IE Integral Neural Networks Reviewed International journal

Honda A., Kamata Y., James S.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 13408 LNAI 168 - 180 2022.08

Zero Additivization of a Monotone Measure on a Finite Set Reviewed International journal

Ryoji Fukuda, Aoi Honda, Yoshiaki Okazaki

Proceedings The 19th International Conference on Modeling Decisions for Artificial Intelligence 2022.08

Non-discrete k-order additivity of a set function and distorted measure Reviewed International journal

Honda A., Fukuda R., Okazaki Y.

Fuzzy Sets and Systems 430 36 - 47 2022.02

Non Additive Set Functions and their Applications—VI Invited Reviewed

Fujimoto Katsushige, Takahagi Eiichiro, Narukawa Yasuo, Honda Aoi

SYSTEMS, CONTROL AND INFORMATION ( THE INSTITUTE OF SYSTEMS, CONTROL AND INFORMATION ENGINEERS ) 65 ( 12 ) 500 - 506 2021.12

　PREV- NEXT 　

<HONDA Aoi>

Papers - HONDA Aoi