Hcore-Init: Neural Network Initialization based on Graph Degeneracy
This work addresses the problem of improving neural network training efficiency for researchers and practitioners by offering a novel initialization technique, though it appears incremental as it builds on existing graph mining concepts.
The paper tackled the problem of neural network initialization by proposing a graph-based method called k-hypercore decomposition, which uses hypergraph degeneracy to re-initialize weights after short pretraining, and it outperformed state-of-the-art initialization methods in experiments on convolutional neural networks and multilayer perceptrons for image recognition tasks.
Neural networks are the pinnacle of Artificial Intelligence, as in recent years we witnessed many novel architectures, learning and optimization techniques for deep learning. Capitalizing on the fact that neural networks inherently constitute multipartite graphs among neuron layers, we aim to analyze directly their structure to extract meaningful information that can improve the learning process. To our knowledge graph mining techniques for enhancing learning in neural networks have not been thoroughly investigated. In this paper we propose an adapted version of the k-core structure for the complete weighted multipartite graph extracted from a deep learning architecture. As a multipartite graph is a combination of bipartite graphs, that are in turn the incidence graphs of hypergraphs, we design k-hypercore decomposition, the hypergraph analogue of k-core degeneracy. We applied k-hypercore to several neural network architectures, more specifically to convolutional neural networks and multilayer perceptrons for image recognition tasks after a very short pretraining. Then we used the information provided by the hypercore numbers of the neurons to re-initialize the weights of the neural network, thus biasing the gradient optimization scheme. Extensive experiments proved that k-hypercore outperforms the state-of-the-art initialization methods.