Graph-Structured Hyperdimensional Computing for Data-Efficient and Explainable Process-Structure-Property Prediction

2026-05-08

AI Summary

This work addresses the challenge of predicting process–structure–property (PSP) relationships in multiphoton photoreduction manufacturing, which is hindered by sparse, heterogeneous, and highly interactive data. To this end, the authors propose PSP-HDC, a novel graph-structured hyperdimensional computing framework that uniquely integrates graph representations with hyperdimensional computing. Leveraging a directed PSP graph as prior knowledge, the method employs a learnable scalar-to-hypervector encoder to handle heterogeneous parameters and performs binding and bundling operations along graph dependencies, enabling unified prediction and interpretation through associative memory. The model offers intrinsic interpretability at the parameter, group, and intra-group levels and quantifies prototype formation via memory alignment and separation. Evaluated on 3D platform sheet resistance interval prediction, PSP-HDC achieves a random-split accuracy of 0.910 ± 0.077 and a process-fold generalization performance of 0.896, significantly outperforming strong baseline models.

Abstract

Multiphoton photoreduction enables high-fidelity fabrication of complex 3D microstructures, yet reliable process-structure-property (PSP) prediction remains difficult because the available data are sparse, heterogeneous, and interaction-dominated. In this regime, conventional feature-vector models are statistically underdetermined, making them prone to spurious correlations, poor regime transfer, and unstable post hoc explanations, whereas mechanistic pipelines depend on calibrated submodels that are rarely available during early process development. We present PSP-HDC, a graph-structured hyperdimensional computing framework that encodes a directed PSP graph as an internal prior for representation, inference, and explanation. A trainable scalar-to-hypervector encoder learns parameter-specific embeddings on a fixed hyperdimensional basis to accommodate heterogeneous scales and noise. Sample representations are then composed through graph-aligned binding and bundling along directed PSP dependencies, and prediction is performed by associative-memory retrieval against class prototypes. Because the same prototype memories support both decision making and attribution, PSP-HDC provides intrinsic explanations at the parameter, group, and within-group levels, while memory alignment and separation quantify prototype formation during training. On sheet-resistance regime prediction for the 3D platform, PSP-HDC achieves an accuracy of 0.910 +/- 0.077 over 1000 random splits and 0.896 under process-fold generalization, outperforming strong baselines.

Problem

• process-structure-property prediction
• data sparsity
• heterogeneous data
• interaction-dominated systems
• multiphoton photoreduction

Innovation

• Hyperdimensional Computing
• Graph-Structured Representation
• Process-Structure-Property Prediction
• Intrinsic Explainability
• Associative Memory