KANGURA: Kolmogorov-Arnold Network-Based Geometry-Aware Learning with Unified Representation Attention for 3D Modeling of Complex Structures

2025-11-17

AI Summary

To address the challenge of capturing geometry-dependent features in complex 3D structures—such as microbial fuel cell (MFC) anodes—this paper proposes KANGURA, the first framework to integrate Kolmogorov–Arnold Networks (KANs) into 3D geometric learning. Departing from conventional multilayer perceptrons (MLPs), KANGURA leverages functional decomposition for geometry-aware representation learning. It further introduces a geometric-decoupled representation scheme and a unified attention mechanism, enabling dynamic enhancement of critical structural regions and facilitating interpretable modeling. Evaluated on ModelNet40, KANGURA achieves 92.7% classification accuracy, outperforming 15 state-of-the-art methods. In practical application to MFC anode performance prediction, it attains 97% accuracy—demonstrating substantial improvements in modeling fidelity and engineering applicability for intricate 3D architectures.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: 3D Computer Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Microbial Fuel Cells (MFCs) offer a promising pathway for sustainable energy generation by converting organic matter into electricity through microbial processes. A key factor influencing MFC performance is the anode structure, where design and material properties play a crucial role. Existing predictive models struggle to capture the complex geometric dependencies necessary to optimize these structures. To solve this problem, we propose KANGURA: Kolmogorov-Arnold Network-Based Geometry-Aware Learning with Unified Representation Attention. KANGURA introduces a new approach to three-dimensional (3D) machine learning modeling. It formulates prediction as a function decomposition problem, where Kolmogorov-Arnold Network (KAN)- based representation learning reconstructs geometric relationships without a conventional multi- layer perceptron (MLP). To refine spatial understanding, geometry-disentangled representation learning separates structural variations into interpretable components, while unified attention mechanisms dynamically enhance critical geometric regions. Experimental results demonstrate that KANGURA outperforms over 15 state-of-the-art (SOTA) models on the ModelNet40 benchmark dataset, achieving 92.7% accuracy, and excels in a real-world MFC anode structure problem with 97% accuracy. This establishes KANGURA as a robust framework for 3D geometric modeling, unlocking new possibilities for optimizing complex structures in advanced manufacturing and quality-driven engineering applications.

Problem

• Existing models struggle to capture complex geometric dependencies for structure optimization
• Predictive models fail to represent 3D geometric relationships in anode structures effectively
• Current approaches cannot adequately model spatial variations in complex microbial fuel cell structures

Innovation

• Kolmogorov-Arnold Network for 3D geometric modeling
• Geometry-disentangled representation learning for interpretability
• Unified attention mechanisms to enhance critical regions