Liang Zhao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Liang Zhao has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
Hiroshi Nagamochi
Department of Applied Mathematics and Physics, Graduate School of Informatics, Kyoto University, Kyoto, Japan
14 shared publications- Tatsuya Akutsu
Bioinformatics Center, Institute for Chemical Research, Kyoto University, Kyoto, Japan
11 shared publications - Kazuya Haraguchi
Kyoto University
8 shared publications - Naveed Ahmed Azam
Department of Mathematics, Quaid-i-Azam University, Islamabad, Pakistan
8 shared publications - Jianshen Zhu
Department of Applied Mathematics and Physics, Graduate School of Informatics, Kyoto University, Kyoto, Japan
7 shared publications Toshihide Ibaraki
7 shared publications
Latest Publications
- 16 November 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 131
Mask to marker: Explainable AI for spatiotemporal analysis of the niacin skin-flush response in mental health screening
- 16 November 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 131
Mask to marker: Explainable AI for spatiotemporal analysis of the niacin skin-flush response in mental health screening
- 19 October 2026
- Published by Taylor & Francis in International Journal of Pavement Engineering
- Vol. 27 (1)
Deployment-oriented benchmarking of YOLO detectors for multiclass pavement distress inspection using street-view imagery
1. The durability and serviceability of pavement infrastructure are essential for safe, efficient, and sustainable transportation systems. Pavement surface distresses, including cracking, potholes,...
- 19 October 2026
- Published by Taylor & Francis in International Journal of Pavement Engineering
- Vol. 27 (1)
Deployment-oriented benchmarking of YOLO detectors for multiclass pavement distress inspection using street-view imagery
1. The durability and serviceability of pavement infrastructure are essential for safe, efficient, and sustainable transportation systems. Pavement surface distresses, including cracking, potholes,...
- Open Access
- 17 February 2026
- Published by Elsevier in SSRN Electronic Journal
Sublinear Scaling in Subnational Parliaments: Empirical Evidence from G20 Countries
This paper examines the relationship between population size and the size of subnational legislatures, a question that has received far less attention than its national-level counterpart despite its importance for political representation. While a substantial literature shows that national parliaments tend to scale sublinearly with population—most commonly between square-root and cubic-root laws—little is known about whether similar patterns hold below the national level. Using an original dataset covering 804 subnational units across the nineteen sovereign G20 countries, including overseas, special, and outlying territories, this study analyzes the scaling relationship between population and the number of legislative seats in the highest subnational institutions. The sample spans a wide diversity of political systems, including federal and unitary states as well as democratic, hybrid, and authoritarian regimes. We estimate two complementary models, a log-linear specification and a nonlinear raw-data specification, to evaluate for sublinear scaling without imposing strong prior assumptions. The results show that proportional scaling has been abandoned at the subnational level and that sublinear scaling dominates across countries. Most countries exhibit scaling coefficients between 0.3 and 0.5, consistent with findings for national parliaments. These results document a first-order institutional regularity at the subnational level, with direct implications for the political economy of legislative design and the governance of population change. We discuss the convergence of scaling behavior across heterogeneous political systems and its implications for institutional reform.
- 17 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSublinear Scaling in Subnational Parliaments: Empirical Evidence from G20 Countries
This paper examines the relationship between population size and the size of subnational legislatures, a question that has received far less attention than its national-level counterpart despite its importance for political representation. While a substantial literature shows that national parliaments tend to scale sublinearly with population—most commonly between square-root and cubic-root laws—little is known about whether similar patterns hold below the national level. Using an original dataset covering 804 subnational units across the nineteen sovereign G20 countries, including overseas, special, and outlying territories, this study analyzes the scaling relationship between population and the number of legislative seats in the highest subnational institutions. The sample spans a wide diversity of political systems, including federal and unitary states as well as democratic, hybrid, and authoritarian regimes. We estimate two complementary models, a log-linear specification and a nonlinear raw-data specification, to evaluate for sublinear scaling without imposing strong prior assumptions. The results show that proportional scaling has been abandoned at the subnational level and that sublinear scaling dominates across countries. Most countries exhibit scaling coefficients between 0.3 and 0.5, consistent with findings for national parliaments. These results document a first-order institutional regularity at the subnational level, with direct implications for the political economy of legislative design and the governance of population change. We discuss the convergence of scaling behavior across heterogeneous political systems and its implications for institutional reform.
- Open Access
- 21 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Combining Graph Neural Networks and Mixed Integer Linear Programming for Molecular Inference under the Two-Layered Model
- 21 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessCombining Graph Neural Networks and Mixed Integer Linear Programming for Molecular Inference under the Two-Layered Model
- Open Access
- 27 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 57-62
Cross-Attentive Prototypical Few-Shot Learning for Early Diagnosis of AMD and Glaucoma from Volumetric OCT Data
Early diagnosis of glaucoma and age-related macular degeneration (AMD) from volumetric optical coherence tomography (OCT) is critical for preventing vision loss, yet current approaches face challenges including time-consuming manual review, inter-reader variability, and scarcity of labeled data. This work introduces CAP-FSL (Cross-Attentive Prototypical Few-Shot Learning), a framework designed for 3D OCT analysis that maintains strong diagnostic performance with only a handful of labeled volumes. Unlike standard Prototypical Networks that form static support-only centroids, CAP-FSL employs query → support cross-attention to dynamically condition support embeddings before prototype formation, yielding query-aware class exemplars that better capture subtle pathological variations. The architecture integrates a 3D ResNet-50 backbone with squeeze-and-excitation (SE) attention to effectively capture volumetric structural patterns while remaining label-efficient and quick to adapt. Evaluated on two public OCT cohorts—one focusing on optic nerve head (ONH) for glaucoma detection and another on macular scans for AMD diagnosis—CAP-FSL consistently outperforms established metric-learning baselines including Matching Networks, standard Prototypical Networks, and Deep Subspace Networks in both 1-shot and 5-shot scenarios. In 5-shot testing, the method achieves classification accuracies of 95.74% ± 2.18% for glaucoma, 93.85% ± 2.70% for AMD, and 92.73% ± 3.18% for normal cases, representing substantial improvements over baseline Prototypical Networks. High sensitivity and specificity metrics with low variance across ten random seeds demonstrate the method’s reliability under data scarcity conditions. The cross-attention mechanism produces interpretable attention maps that align with known disease biomarkers, enhancing clinical interpretability. By combining query-conditioned prototypes with efficient 3D feature extraction, CAP-FSL offers a practical approach for rapidly adapting OCT screening and triage models in data-constrained clinical settings. Code and trained models are publicly available at https://github.com/Rimsa78/CAP-FSL.
- 27 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 57-62
Open AccessCross-Attentive Prototypical Few-Shot Learning for Early Diagnosis of AMD and Glaucoma from Volumetric OCT Data
Early diagnosis of glaucoma and age-related macular degeneration (AMD) from volumetric optical coherence tomography (OCT) is critical for preventing vision loss, yet current approaches face challenges including time-consuming manual review, inter-reader variability, and scarcity of labeled data. This work introduces CAP-FSL (Cross-Attentive Prototypical Few-Shot Learning), a framework designed for 3D OCT analysis that maintains strong diagnostic performance with only a handful of labeled volumes. Unlike standard Prototypical Networks that form static support-only centroids, CAP-FSL employs query → support cross-attention to dynamically condition support embeddings before prototype formation, yielding query-aware class exemplars that better capture subtle pathological variations. The architecture integrates a 3D ResNet-50 backbone with squeeze-and-excitation (SE) attention to effectively capture volumetric structural patterns while remaining label-efficient and quick to adapt. Evaluated on two public OCT cohorts—one focusing on optic nerve head (ONH) for glaucoma detection and another on macular scans for AMD diagnosis—CAP-FSL consistently outperforms established metric-learning baselines including Matching Networks, standard Prototypical Networks, and Deep Subspace Networks in both 1-shot and 5-shot scenarios. In 5-shot testing, the method achieves classification accuracies of 95.74% ± 2.18% for glaucoma, 93.85% ± 2.70% for AMD, and 92.73% ± 3.18% for normal cases, representing substantial improvements over baseline Prototypical Networks. High sensitivity and specificity metrics with low variance across ten random seeds demonstrate the method’s reliability under data scarcity conditions. The cross-attention mechanism produces interpretable attention maps that align with known disease biomarkers, enhancing clinical interpretability. By combining query-conditioned prototypes with efficient 3D feature extraction, CAP-FSL offers a practical approach for rapidly adapting OCT screening and triage models in data-constrained clinical settings. Code and trained models are publicly available at https://github.com/Rimsa78/CAP-FSL.
- Open Access
- 26 March 2025
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 17 (1) , 1-15
A unified approach to inferring chemical compounds with the desired aqueous solubility
Aqueous solubility (AS) is a key physiochemical property that plays a crucial role in drug discovery and material design. We report a novel unified approach to predict and infer chemical compounds with the desired AS based on simple deterministic graph-theoretic descriptors, multiple linear regression (MLR), and mixed integer linear programming (MILP). Selected descriptors based on a forward stepwise procedure enabled the simplest regression model, MLR, to achieve significantly good prediction accuracy compared to the existing approaches, achieving accuracy in the range [0.7191, 0.9377] for 29 diverse datasets. By simulating these descriptors and learning models as MILPs, we inferred mathematically exact and optimal compounds with the desired AS, prescribed structures, and up to 50 non-hydrogen atoms in a reasonable time range [6, 1166] seconds. These findings indicate a strong correlation between the simple graph-theoretic descriptors and the AS of compounds, potentially leading to a deeper understanding of their AS without relying on widely used complicated chemical descriptors and complex machine learning models that are computationally expensive, and therefore difficult to use for inference. An implementation of the proposed approach is available at https://github.com/ku-dml/mol-infer/tree/master/AqSol.
- 26 March 2025
- Published by Springer Nature in Journal of Cheminformatics
- Vol. 17 (1) , 1-15
Open AccessA unified approach to inferring chemical compounds with the desired aqueous solubility
Aqueous solubility (AS) is a key physiochemical property that plays a crucial role in drug discovery and material design. We report a novel unified approach to predict and infer chemical compounds with the desired AS based on simple deterministic graph-theoretic descriptors, multiple linear regression (MLR), and mixed integer linear programming (MILP). Selected descriptors based on a forward stepwise procedure enabled the simplest regression model, MLR, to achieve significantly good prediction accuracy compared to the existing approaches, achieving accuracy in the range [0.7191, 0.9377] for 29 diverse datasets. By simulating these descriptors and learning models as MILPs, we inferred mathematically exact and optimal compounds with the desired AS, prescribed structures, and up to 50 non-hydrogen atoms in a reasonable time range [6, 1166] seconds. These findings indicate a strong correlation between the simple graph-theoretic descriptors and the AS of compounds, potentially leading to a deeper understanding of their AS without relying on widely used complicated chemical descriptors and complex machine learning models that are computationally expensive, and therefore difficult to use for inference. An implementation of the proposed approach is available at https://github.com/ku-dml/mol-infer/tree/master/AqSol.
- Open Access
A Unified Approach to Inferring Chemical Compounds with the Desired Aqueous Solubility
Aqueous solubility (AS) is a key physiochemical property that plays a crucial role in drug discovery and material design. We report a novel unified approach to predict and infer chemical compounds with the desired AS based on simple deterministic graph-theoretic descriptors, multiple linear regression (MLR) and mixed integer linear programming (MILP). Selected descriptors based on a forward stepwise procedure enabled the simplest regression model, MLR, to achieve significantly good prediction accuracy compared to the existing approaches, achieving the accuracy in the range [0.7191, 0.9377] for 29 diverse datasets. By simulating these descriptors and learning models as MILPs, we inferred mathematically exact and optimal compounds with the desired AS, prescribed structures, and up to 50 non-hydrogen atoms in a reasonable time range [6, 1204] seconds. These findings indicate a strong correlation between the simple graph-theoretic descriptors and the AS of compounds, potentially leading to a deeper understanding of their AS without relying on widely used complicated chemical descriptors and complex machine learning models that are computationally expensive, and therefore difficult to use for inference. An implementation of the proposed approach is available at https://github.com/ku-dml/mol-infer/tree/master/AqSol.
Open AccessA Unified Approach to Inferring Chemical Compounds with the Desired Aqueous Solubility
Aqueous solubility (AS) is a key physiochemical property that plays a crucial role in drug discovery and material design. We report a novel unified approach to predict and infer chemical compounds with the desired AS based on simple deterministic graph-theoretic descriptors, multiple linear regression (MLR) and mixed integer linear programming (MILP). Selected descriptors based on a forward stepwise procedure enabled the simplest regression model, MLR, to achieve significantly good prediction accuracy compared to the existing approaches, achieving the accuracy in the range [0.7191, 0.9377] for 29 diverse datasets. By simulating these descriptors and learning models as MILPs, we inferred mathematically exact and optimal compounds with the desired AS, prescribed structures, and up to 50 non-hydrogen atoms in a reasonable time range [6, 1204] seconds. These findings indicate a strong correlation between the simple graph-theoretic descriptors and the AS of compounds, potentially leading to a deeper understanding of their AS without relying on widely used complicated chemical descriptors and complex machine learning models that are computationally expensive, and therefore difficult to use for inference. An implementation of the proposed approach is available at https://github.com/ku-dml/mol-infer/tree/master/AqSol.
- 22 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Computational Biology and Bioinformatics
- Vol. 21 (6) , 1623-1632
A Method for Inferring Polymers Based on Linear Regression and Integer Programming
A novel framework has recently been proposed for designing the molecular structure of chemical compounds with a desired chemical property using both artificial neural networks and mixed integer linear programming. In this paper, we design a new method for inferring a polymer based on the framework. For this, we introduce a new way of representing a polymer as a form of monomer and define new descriptors that feature the structure of polymers. We also use linear regression as a building block of constructing a prediction function in the framework. The results of our computational experiments reveal a set of chemical properties on polymers to which a prediction function constructed with linear regression performs well. We also observe that the proposed method can infer polymers with up to 50 nonhydrogen atoms in a monomer form
- 22 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Computational Biology and Bioinformatics
- Vol. 21 (6) , 1623-1632
A Method for Inferring Polymers Based on Linear Regression and Integer Programming
A novel framework has recently been proposed for designing the molecular structure of chemical compounds with a desired chemical property using both artificial neural networks and mixed integer linear programming. In this paper, we design a new method for inferring a polymer based on the framework. For this, we introduce a new way of representing a polymer as a form of monomer and define new descriptors that feature the structure of polymers. We also use linear regression as a building block of constructing a prediction function in the framework. The results of our computational experiments reveal a set of chemical properties on polymers to which a prediction function constructed with linear regression performs well. We also observe that the proposed method can infer polymers with up to 50 nonhydrogen atoms in a monomer form
- 2 July 2024
- Published by Springer Nature in Social Network Analysis and Mining
- Vol. 14 (1) , 1-13
On a pyramid structure in social networks
- 2 July 2024
- Published by Springer Nature in Social Network Analysis and Mining
- Vol. 14 (1) , 1-13
On a pyramid structure in social networks
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Computational Biology and Bioinformatics
- Vol. 21 (5) , 1529-1541
Molecular Design Based on Integer Programming and Splitting Data Sets by Hyperplanes
A novel framework for designing the molecular structure of chemical compounds with a desired chemical property has recently been proposed. The framework infers a desired chemical graph by solving a mixed integer linear program (MILP) that simulates the computation process of two functions: a feature function defined by a two-layered model on chemical graphs and a prediction function constructed by a machine learning method. To improve the learning performance of prediction functions in the framework, we design a method that splits a given data set $\mathcal {C}$ into two subsets $\mathcal {C}^{(i)},i=1,2$ by a hyperplane in a chemical space so that most compounds in the first (resp., second) subset have observed values lower (resp., higher) than a threshold $\theta$ . We construct a prediction function $\psi$ to the data set $\mathcal {C}$ by combining prediction functions $\psi _{i},i=1,2$ each of which is constructed on $\mathcal {C}^{(i)}$ independently. The results of our computational experiments suggest that the proposed method improved the learning performance for several chemical properties to which a good prediction function has been difficult to construct.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Computational Biology and Bioinformatics
- Vol. 21 (5) , 1529-1541
Molecular Design Based on Integer Programming and Splitting Data Sets by Hyperplanes
A novel framework for designing the molecular structure of chemical compounds with a desired chemical property has recently been proposed. The framework infers a desired chemical graph by solving a mixed integer linear program (MILP) that simulates the computation process of two functions: a feature function defined by a two-layered model on chemical graphs and a prediction function constructed by a machine learning method. To improve the learning performance of prediction functions in the framework, we design a method that splits a given data set $\mathcal {C}$ into two subsets $\mathcal {C}^{(i)},i=1,2$ by a hyperplane in a chemical space so that most compounds in the first (resp., second) subset have observed values lower (resp., higher) than a threshold $\theta$ . We construct a prediction function $\psi$ to the data set $\mathcal {C}$ by combining prediction functions $\psi _{i},i=1,2$ each of which is constructed on $\mathcal {C}^{(i)}$ independently. The results of our computational experiments suggest that the proposed method improved the learning performance for several chemical properties to which a good prediction function has been difficult to construct.