AutoHGNN: Robust and efficient neural architecture search for hypergraph neural networks
This paper proposes AutoHGNN, a neural architecture search framework tailored for hypergraph neural networks, and introduces a Hyper-Interaction Module (HIM) into the search space to address the mismatch between conventional graph neural network designs and hypergraph data.
Sirui Li, Pietro Lio', Xin-Sheng Li et al.
· Knowledge-Based Systems · 0 citations