Search papers, labs, and topics across Lattice.
The paper introduces FI-LDP-HGAT, a framework for privacy-preserving graph representation learning in metal additive manufacturing that combines a hierarchical graph attention network (HGAT) to model spatial dependencies with a feature-importance-aware anisotropic Gaussian mechanism (FI-LDP) for privatization. FI-LDP strategically allocates the privacy budget by adding less noise to important thermal features and more to redundant ones, unlike isotropic LDP which applies uniform noise. Experiments on a DED porosity dataset show FI-LDP-HGAT significantly improves utility recovery and defect recall compared to standard methods under various privacy budgets, demonstrating the effectiveness of anisotropic noise allocation.
Shielding metal additive manufacturing data doesn't have to mean sacrificing accuracy: prioritizing privacy for less-critical features preserves model performance while ensuring data security.
Metal additive manufacturing (AM) enables the fabrication of safety-critical components, but reliable quality assurance depends on high-fidelity sensor streams containing proprietary process information, limiting collaborative data sharing. Existing defect-detection models typically treat melt-pool observations as independent samples, ignoring layer-wise physical couplings. Moreover, conventional privacy-preserving techniques, particularly Local Differential Privacy (LDP), lead to severe utility degradation because they inject uniform noise across all feature dimensions. To address these interrelated challenges, we propose FI-LDP-HGAT. This computational framework combines two methodological components: a stratified Hierarchical Graph Attention Network (HGAT) that captures spatial and thermal dependencies across scan tracks and deposited layers, and a feature-importance-aware anisotropic Gaussian mechanism (FI-LDP) for non-interactive feature privatization. Unlike isotropic LDP, FI-LDP redistributes the privacy budget across embedding coordinates using an encoder-derived importance prior, assigning lower noise to task-critical thermal signatures and higher noise to redundant dimensions while maintaining formal LDP guarantees. Experiments on a Directed Energy Deposition (DED) porosity dataset demonstrate that FI-LDP-HGAT achieves 81.5% utility recovery at a moderate privacy budget (epsilon = 4) and maintains defect recall of 0.762 under strict privacy (epsilon = 2), while outperforming classical ML, standard GNNs, and alternative privacy mechanisms, including DP-SGD across all evaluated metrics. Mechanistic analysis confirms a strong negative correlation (Spearman = -0.81) between feature importance and noise magnitude, providing interpretable evidence that the privacy-utility gains are driven by principled anisotropic allocation.
