Penguin Parallel Packed Encryption for Fast Graph Network Inference

HE operations (e.g., ciphertext (ct) rotations/multiplications, additions), which could be orders of For example, a GCN layer's computation is dominated by the special consecutive HE operations are defined in Sec. 2. For generality, we assume both feature matrix and adjacency Parallel-Packing (see Sec. 3.2), the ciphertext size is fully exploited, and the total HE operation count GCN layer's output feature matrix H. Throughout this work, we refer A as the normalized adjacency We adopt a threat model setting consistent with prior works [9, 14, 3, 7, 18, 22, 27]. The cloud server is semi-honest (e.g.