HDReason: Algorithm-Hardware Codesign for Hyperdimensional Knowledge Graph Reasoning

arXiv:2403.05763v1 Announce Type: cross
Abstract: In recent times, a plethora of hardware accelerators have been put forth for graph learning applications such as vertex classification and graph classification. However, previous works have paid little attention to Knowledge Graph Completion (KGC), a task that is well-known for its significantly higher algorithm complexity. The state-of-the-art KGC solutions based on graph convolution neural network (GCN) involve extensive vertex/relation embedding updates and complicated score functions, which are inherently cumbersome for acceleration. As a result, …

abstract accelerators algorithm applications arxiv attention classification complexity cs.ai cs.ar cs.lg graph graph learning hardware however knowledge knowledge graph reasoning type vertex