Connectivity Matters: Neural Network Pruning Through the Lens of Effective Sparsity

Abstract

Neural network pruning is a fruitful area of research with surging interest in high sparsity regimes. Benchmarking in this domain heavily relies on faithful representation of the sparsity of subnetworks, which has been traditionally computed as the fraction of removed connections (direct sparsity). This definition, however, fails to recognize unpruned parameters that detached from input or output layers of underlying subnetworks, potentially underestimating actual effective sparsity: the fraction of inactivated connections. While this effect might be negligible for moderately pruned networks (up to 10-100 compression rates), we find that it plays an increasing role for thinner subnetworks, greatly distorting comparison between different pruning algorithms. For example, we show that effective compression of a randomly pruned LeNet-300-100 can be orders of magnitude larger than its direct counterpart, while no discrepancy is ever observed when using SynFlow for pruning [Tanaka et al., 2020]. In this work, we adopt the lens of effective sparsity to reevaluate several recent pruning algorithms on common benchmark architectures (e.g., LeNet-300-100, VGG-19, ResNet-18) and discover that their absolute and relative performance changes dramatically in this new and more appropriate framework. To aim for effective, rather than direct, sparsity, we develop a low-cost extension to most pruning algorithms. Further, equipped with effective sparsity as a reference frame, we partially reconfirm that random pruning with appropriate sparsity allocation across layers performs as well or better than more sophisticated algorithms for pruning at initialization [Su et al., 2020]. In response to this observation, using a simple analogy of pressure distribution in coupled cylinders from physics, we design novel layerwise sparsity quotas that outperform all existing baselines in the context of random pruning.

References in corpus (14)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Distilling the Knowledge in a Neural Network
• Compressing Deep Convolutional Networks using Vector Quantization
• Rethinking the Value of Network Pruning
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• Compressing Neural Networks with the Hashing Trick
• SNIP: Single-shot Network Pruning based on Connection Sensitivity
• Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Science
• The State of Sparsity in Deep Neural Networks
• Variational Dropout Sparsifies Deep Neural Networks
• Sparse Networks from Scratch: Faster Training without Losing Performance
• Picking Winning Tickets Before Training by Preserving Gradient Flow
• Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterization
• Layer-adaptive sparsity for the Magnitude-based Pruning