ATAkihiro Takemura

Akihiro Takemura

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Akihiro Takemura has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Formally Explaining Decision Tree Models with Answer Set Programming

    by Akihiro Takemura, Masayuki Otani, Katsumi Inoue

    Decision tree models, including random forests and gradient-boosted decision trees, are widely used in machine learning due to their high predictive performance. However, their complex structures often make them difficult to interpret, especially in safety-critical applications where model decisions require formal justification. Recent work has demonstrated that logical and abductive explanations can be derived through automated reasoning techniques. In this paper, we propose a method for generating various types of explanations, namely, sufficient, contrastive, majority, and tree-specific explanations, using Answer Set Programming (ASP). Compared to SAT-based approaches, our ASP-based method offers greater flexibility in encoding user preferences and supports enumeration of all possible explanations. We empirically evaluate the approach on a diverse set of datasets and demonstrate its effectiveness and limitations compared to existing methods.


  • Open Access

    Towards End-to-End ASP Computation

    by Taisuke Sato, Akihiro Takemura, Katsumi Inoue

    We propose an end-to-end approach for Answer Set Programming (ASP) and linear algebraically compute stable models satisfying given constraints. The idea is to implement Lin-Zhao’s theorem together with constraints directly in vector spaces as numerical minimization of a cost function constructed from a matricized normal logic program, loop formulas in Lin-Zhao’s theorem and constraints, thereby no use of symbolic ASP or SAT solvers involved in our approach. We also propose precomputation that shrinks the program size and heuristics for loop formulas to reduce computational difficulty. We empirically test our approach with programming examples including the three-coloring and Hamiltonian cycle problems.


  • Open Access

    Differentiable Logic Programming for Distant Supervision

    by Akihiro Takemura, Katsumi Inoue

    We introduce a new method for integrating neural networks with logic programming in Neural-Symbolic AI (NeSy), aimed at learning with distant supervision, in which direct labels are unavailable. Unlike prior methods, our approach does not depend on symbolic solvers for reasoning about missing labels. Instead, it evaluates logical implications and constraints in a differentiable manner by embedding both neural network outputs and logic programs into matrices. This method facilitates more efficient learning under distant supervision. We evaluated our approach against existing methods while maintaining a constant volume of training data. The findings indicate that our method not only matches or exceeds the accuracy of other methods across various tasks but also speeds up the learning process. These results highlight the potential of our approach to enhance both accuracy and learning efficiency in NeSy applications.


  • Open Access

    Generating Global and Local Explanations for Tree-Ensemble Learning Methods by Answer Set Programming

    by Akihiro Takemura, Katsumi Inoue

    We propose a method for generating rule sets as global and local explanations for tree-ensemble learning methods using Answer Set Programming (ASP). To this end, we adopt a decompositional approach where the split structures of the base decision trees are exploited in the construction of rules, which in turn are assessed using pattern mining methods encoded in ASP to extract explanatory rules. For global explanations, candidate rules are chosen from the entire trained tree-ensemble models, whereas for local explanations, candidate rules are selected by only considering rules that are relevant to the particular predicted instance. We show how user-defined constraints and preferences can be represented declaratively in ASP to allow for transparent and flexible rule set generation, and how rules can be used as explanations to help the user better understand the models. Experimental evaluation with real-world datasets and popular tree-ensemble algorithms demonstrates that our approach is applicable to a wide range of classification tasks. Under consideration in Theory and Practice of Logic Programming (TPLP).

  • Open Access

    Generating Global and Local Explanations for Tree-Ensemble Learning Methods by Answer Set Programming

    by Akihiro Takemura, Katsumi Inoue

    We propose a method for generating rule sets as global and local explanations for tree-ensemble learning methods using answer set programming (ASP). To this end, we adopt a decompositional approach where the split structures of the base decision trees are exploited in the construction of rules, which in turn are assessed using pattern mining methods encoded in ASP to extract explanatory rules. For global explanations, candidate rules are chosen from the entire trained tree-ensemble models, whereas for local explanations, candidate rules are selected by only considering rules that are relevant to the particular predicted instance. We show how user-defined constraints and preferences can be represented declaratively in ASP to allow for transparent and flexible rule set generation, and how rules can be used as explanations to help the user better understand the models. Experimental evaluation with real-world datasets and popular tree-ensemble algorithms demonstrates that our approach is applicable to a wide range of classification tasks.


  • Open Access

    Differentiable Logic Programming for Distant Supervision

    by Akihiro Takemura, Katsumi Inoue

    We introduce a new method for integrating neural networks with logic programming in Neural-Symbolic AI (NeSy), aimed at learning with distant supervision, in which direct labels are unavailable. Unlike prior methods, our approach does not depend on symbolic solvers for reasoning about missing labels. Instead, it evaluates logical implications and constraints in a differentiable manner by embedding both neural network outputs and logic programs into matrices. This method facilitates more efficient learning under distant supervision. We evaluated our approach against existing methods while maintaining a constant volume of training data. The findings indicate that our method not only matches or exceeds the accuracy of other methods across various tasks but also speeds up the learning process. These results highlight the potential of our approach to enhance both accuracy and learning efficiency in NeSy applications.

  • Open Access

    MOD-CL: Multi-label Object Detection with Constrained Loss

    by Sota Moriyama, Koji Watanabe, Katsumi Inoue, Akihiro Takemura

    We introduce MOD-CL, a multi-label object detection framework that utilizes constrained loss in the training process to produce outputs that better satisfy the given requirements. In this paper, we use $\mathrm{MOD_{YOLO}}$, a multi-label object detection model built upon the state-of-the-art object detection model YOLOv8, which has been published in recent years. In Task 1, we introduce the Corrector Model and Blender Model, two new models that follow after the object detection process, aiming to generate a more constrained output. For Task 2, constrained losses have been incorporated into the $\mathrm{MOD_{YOLO}}$ architecture using Product T-Norm. The results show that these implementations are instrumental to improving the scores for both Task 1 and Task 2.


  • Open Access

    Halogen‐ and Hydrogen‐Bonded Salts and Co‐crystals Formed from 4‐Halo‐2,3,5,6‐tetrafluorophenol and Cyclic Secondary and Tertiary Amines: Orthogonal and Non‐orthogonal Halogen and Hydrogen Bonding, and Synthetic Analogues of Halogen‐Bonded Biological Systems

    by Akihiro Takemura, Linda J. McAllister, Sam Hart, Natalie E. Pridmore, Peter B. Karadakov, Adrian C. Whitwood, Duncan W. Bruce

    Co‐crystallisation of, in particular, 4‐iodotetrafluorophenol with a series of secondary and tertiary cyclic amines results in deprotonation of the phenol and formation of the corresponding ammonium phenate. Careful examination of the X‐ray single‐crystal structures shows that the phenate anion develops a CO double bond and that the CC bond lengths in the ring suggest a Meissenheimer‐like delocalisation. This delocalisation is supported by the geometry of the phenate anion optimised at the MP2(Full) level of theory within the aug‐cc‐pVDZ basis (aug‐cc‐pVDZ‐PP on I) and by natural bond orbital (NBO) analyses. With sp2 hybridisation at the phenate oxygen atom, there is strong preference for the formation of two non‐covalent interactions with the oxygen sp2 lone pairs and, in the case of secondary amines, this occurs through hydrogen bonding to the ammonium hydrogen atoms. However, where tertiary amines are concerned, there are insufficient hydrogen atoms available and so an electrophilic iodine atom from a neighbouring 4‐iodotetrafluorophenate group forms an I⋅⋅⋅O halogen bond to give the second interaction. However, in some co‐crystals with secondary amines, it is also found that in addition to the two hydrogen bonds forming with the phenate oxygen sp2 lone pairs, there is an additional intermolecular I⋅⋅⋅O halogen bond in which the electrophilic iodine atom interacts with the CO π‐system. All attempts to reproduce this behaviour with 4‐bromotetrafluorophenol were unsuccessful. These structural motifs are significant as they reproduce extremely well, in low‐molar‐mass synthetic systems, motifs found by Ho and co‐workers when examining halogen‐bonding interactions in biological systems. The analogy is cemented through the structures of co‐crystals of 1,4‐diiodotetrafluorobenzene with acetamide and with N‐methylbenzamide, which, as designed models, demonstrate the orthogonality of hydrogen and halogen bonding proposed in Ho’s biological study.