🤖 AI Summary
To address the low efficiency of secure decision tree training and inference via Multi-Party Computation (MPC) in privacy-sensitive scenarios, this paper proposes the first GPU-accelerated MPC decision tree framework. Methodologically, it integrates MPC protocols, GPU-optimized parallelism, and oblivious array access to achieve structural privacy under the three-party semi-honest model—revealing only tree depth and dataset size, while hiding split features and traversal paths. Crucially, the MPC protocols are deeply co-designed with GPU architecture to minimize both computational and communication overhead. Experiments demonstrate that training is 11× and 21× faster than SPECT and Adult baselines, respectively; inference on a 7-layer tree with 10⁴ samples achieves a 126× speedup. This work establishes the first high-security, high-efficiency GPU-accelerated MPC decision tree framework.
📝 Abstract
Decision tree (DT) is a widely used machine learning model due to its versatility, speed, and interpretability. However, for privacy-sensitive applications, outsourcing DT training and inference to cloud platforms raise concerns about data privacy. Researchers have developed privacy-preserving approaches for DT training and inference using cryptographic primitives, such as Secure Multi-Party Computation (MPC). While these approaches have shown progress, they still suffer from heavy computation and communication overheads. Few recent works employ Graphical Processing Units (GPU) to improve the performance of MPC-protected deep learning. This raises a natural question: extit{can MPC-protected DT training and inference be accelerated by GPU?} We present GTree, the first scheme that uses GPU to accelerate MPC-protected secure DT training and inference. GTree is built across 3 parties who securely and jointly perform each step of DT training and inference with GPU. Each MPC protocol in GTree is designed in a GPU-friendly version. The performance evaluation shows that GTree achieves ${ hicksim}11{ imes}$ and ${ hicksim}21{ imes}$ improvements in training SPECT and Adult datasets, compared to the prior most efficient CPU-based work. For inference, GTree shows its superior efficiency when the DT has less than 10 levels, which is $126 imes$ faster than the prior most efficient work when inferring $10^4$ instances with a tree of 7 levels. GTree also achieves a stronger security guarantee than prior solutions, which only leaks the tree depth and size of data samples while prior solutions also leak the tree structure. With extit{oblivious array access}, the access pattern on GPU is also protected.
