Training Behavior of Sparse Neural Network Topologies
This work tackles memory constraints in neural network training for AI researchers, though it is incremental as it builds on existing sparse methods.
The paper investigates the training behavior of sparse neural network topologies, including pruning-based and RadiX-Nets, to address memory limitations in deep learning. Results indicate that sparse networks achieve accuracies similar to dense ones, but extreme sparsity leads to training instability.
Improvements in the performance of deep neural networks have often come through the design of larger and more complex networks. As a result, fast memory is a significant limiting factor in our ability to improve network performance. One approach to overcoming this limit is the design of sparse neural networks, which can be both very large and efficiently trained. In this paper we experiment training on sparse neural network topologies. We test pruning-based topologies, which are derived from an initially dense network whose connections are pruned, as well as RadiX-Nets, a class of network topologies with proven connectivity and sparsity properties. Results show that sparse networks obtain accuracies comparable to dense networks, but extreme levels of sparsity cause instability in training, which merits further study.