Miniaturized Graph Convolutional Networks with Topologically Consistent Pruning
This work addresses a specific bottleneck in lightweight architecture design for graph convolutional networks, offering an incremental improvement over existing pruning methods.
The paper tackles the problem of topological inconsistency in magnitude pruning under high pruning regimes, which leads to disconnected subnetworks and poor generalization, by introducing a novel pruning method that ensures topological consistency, resulting in significantly enhanced generalization for skeleton-based action recognition.
Magnitude pruning is one of the mainstream methods in lightweight architecture design whose goal is to extract subnetworks with the largest weight connections. This method is known to be successful, but under very high pruning regimes, it suffers from topological inconsistency which renders the extracted subnetworks disconnected, and this hinders their generalization ability. In this paper, we devise a novel magnitude pruning method that allows extracting subnetworks while guarantying their topological consistency. The latter ensures that only accessible and co-accessible -- impactful -- connections are kept in the resulting lightweight networks. Our solution is based on a novel reparametrization and two supervisory bi-directional networks which implement accessibility/co-accessibility and guarantee that only connected subnetworks will be selected during training. This solution allows enhancing generalization significantly, under very high pruning regimes, as corroborated through extensive experiments, involving graph convolutional networks, on the challenging task of skeleton-based action recognition.