MoE-Enhanced Explainable Deep Manifold Transformation for Complex Data Embedding and Visualization

Abstract

Dimensionality reduction (DR) plays a crucial role in various fields, including data engineering and visualization, by simplifying complex datasets while retaining essential information. However, achieving both high DR accuracy and strong explainability remains a fundamental challenge, especially for users dealing with high-dimensional data. Traditional DR methods often face a trade-off between precision and transparency, where optimizing for performance can lead to reduced explainability, and vice versa. This limitation is especially prominent in real-world applications such as image, tabular, and text data analysis, where both accuracy and explainability are critical. To address these challenges, this work introduces the MoE-based Explainable Deep Manifold Transformation (DMT-ME). The proposed approach combines a geometry-aware hyperbolic mapper with Mixture of Experts (MoE) models, where sparse expert specialization provides the main representational gain and the hyperbolic component offers an additional refinement for structurally complex data. DMT-ME enhances DR accuracy primarily through MoE-based sparse routing and structure-aware matching, while also improving explainability by explicitly linking input data, embedding outcomes, and key features through the MoE structure. Extensive experiments demonstrate that DMT-ME consistently achieves superior performance in both DR accuracy and model explainability, making it a robust solution for complex data analysis. The code is available at https://github.com/zangzelin/code_dmtme.

17 pages, 15 figures, accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)

References in corpus (8)

• Interpretability of machine learning based prediction models in healthcare
• Visualizing Large-scale and High-dimensional Data
• DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection
• DimReader: Axis lines that explain non-linear projections
• Extendable and invertible manifold learning with geometry regularized autoencoders
• Reducing Data Complexity using Autoencoders with Class-informed Loss Functions
• Dimension Reduction with Prior Information for Knowledge Discovery
• Higher-dimensional Euclidean and non-Euclidean structures in planar circuit quantum electrodynamics