arXiv subjects

Hiroki Seto

Publications and source records attributed to Hiroki Seto.

3 recordsLinked to original sources

arXiv2026

Generalized bilinear Koopman realization from input-output data for multi-step prediction with metaheuristic optimization of lifting function and its application to real-world industrial system

This paper introduces an input-output bilinear Koopman realization with an optimization algorithm of lifting functions. For nonlinear systems with inputs, Koopman-based modeling is effective because the Koopman operator enables a high-dimensional linear representation of nonlinear dynamics. However, traditional approaches face significant challenges in industrial applications. Measuring all system states is often impractical due to constraints on sensor installation. Moreover, the predictive performance of a Koopman model strongly depends on the choice of lifting functions, and their design typically requires substantial manual effort. In addition, although a linear time-invariant (LTI) Koopman model is the most commonly used model structure in the Koopman framework, such model exhibit limited predictive accuracy. To address these limitations, we propose an input-output bilinear Koopman modeling in which the design parameters of radial basis function (RBF)-based lifting functions are optimized using a global metaheuristic algorithm to improve long-term prediction performance. Consideration of the long-term prediction performance enhances the reliability of the resulting model. The proposed methodology is validated in simulations and experimental tests, with the airpath control system of a diesel engine as the plant to be modeled. This plant represents a challenging industrial application because it exhibits strong nonlinearities and coupled multi-input multi-output (MIMO) dynamics. These results demonstrate that the proposed input-output bilinear Koopman model significantly outperforms traditional linear Koopman models in predictive accuracy.

eess.SY↗

arXiv2025

Sparse Identification of Nonlinear Dynamics Enhanced by Ensemble Learning, Multi-Step Prediction Evaluation, Elite Strategy, and Classification Techniques for Applications to Industrial Systems

This paper proposes a sparse identification of nonlinear dynamics (SINDy) with control and exogenous inputs for highly accurate and reliable prediction. Although SINDy is recognized as a remarkable approach for identifying nonlinear systems, several challenges remain. Its application to industrial systems remains limited, and multi-step predictions are not guaranteed due to overfitting and noisy data. This phenomenon is often caused by the increase in basis functions resulting from the extension of coordinates, such as time-delay embedding. To address these problems, this study proposes an emphasized SINDy framework by integrating ensemble-learning, multi-step prediction evaluations, elite strategy, and classification techniques (EMEC-SINDy), while preserving convex optimization. The proposed method employs library bagging and extracts elites with an R-squared greater than 90%. Then, clustering is performed on the surviving elites because physically motivated basis functions are not always available, and the elites obtained do not always have similar basis functions. After the classification, discrete model candidates are obtained by taking the mean of each classified elite. Finally, the best model is selected. Simulation results demonstrate that EMEC-SINDy significantly outperforms original SINDy approaches in multi-step prediction accuracy under noisy conditions, validating its applicability to the diesel engine airpath system, which is known as a complex and highly coupled nonlinear multi-input multi-output system.

eess.SY↗

arXiv2024

ProcBench: Benchmark for Multi-Step Reasoning and Following Procedure

Reasoning is central to a wide range of intellectual activities, and while the capabilities of large language models (LLMs) continue to advance, their performance in reasoning tasks remains limited. The processes and mechanisms underlying reasoning are not yet fully understood, but key elements include path exploration, selection of relevant knowledge, and multi-step inference. Problems are solved through the synthesis of these components. In this paper, we propose a benchmark that focuses on a specific aspect of reasoning ability: the direct evaluation of multi-step inference. To this end, we design a special reasoning task where multi-step inference is specifically focused by largely eliminating path exploration and implicit knowledge utilization. Our dataset comprises pairs of explicit instructions and corresponding questions, where the procedures necessary for solving the questions are entirely detailed within the instructions. This setup allows models to solve problems solely by following the provided directives. By constructing problems that require varying numbers of steps to solve and evaluating responses at each step, we enable a thorough assessment of state-of-the-art LLMs' ability to follow instructions. To ensure the robustness of our evaluation, we include multiple distinct tasks. Furthermore, by comparing accuracy across tasks, utilizing step-aware metrics, and applying separately defined measures of complexity, we conduct experiments that offer insights into the capabilities and limitations of LLMs in reasoning tasks. Our findings have significant implications for the development of LLMs and highlight areas for future research in advancing their reasoning abilities. Our dataset is available at \url{https://huggingface.co/datasets/ifujisawa/procbench} and code at \url{https://github.com/ifujisawa/proc-bench}.

cs.AI↗