The Elastic Lottery Ticket Hypothesis

Abstract

Lottery Ticket Hypothesis (LTH) raises keen attention to identifying sparse trainable subnetworks, or winning tickets, which can be trained in isolation to achieve similar or even better performance compared to the full models. Despite many efforts being made, the most effective method to identify such winning tickets is still Iterative Magnitude-based Pruning (IMP), which is computationally expensive and has to be run thoroughly for every different network. A natural question that comes in is: can we "transform" the winning ticket found in one network to another with a different architecture, yielding a winning ticket for the latter at the beginning, without re-doing the expensive IMP? Answering this question is not only practically relevant for efficient "once-for-all" winning ticket finding, but also theoretically appealing for uncovering inherently scalable sparse patterns in networks. We conduct extensive experiments on CIFAR-10 and ImageNet, and propose a variety of strategies to tweak the winning tickets found from different networks of the same model family (e.g., ResNets). Based on these results, we articulate the Elastic Lottery Ticket Hypothesis (E-LTH): by mindfully replicating (or dropping) and re-ordering layers for one network, its corresponding winning ticket could be stretched (or squeezed) into a subnetwork for another deeper (or shallower) network from the same family, whose performance is nearly the same competitive as the latter's winning ticket directly found by IMP. We have also extensively compared E-LTH with pruning-at-initialization and dynamic sparse training methods, as well as discussed the generalizability of E-LTH to different model families, layer types, and across datasets. Code is available at https://github.com/VITA-Group/ElasticLTH.

arXiv commentAccepted at NeurIPS 2021

References in corpus (27)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Neural Tangent Kernel: Convergence and Generalization in Neural Networks
• Rethinking the Value of Network Pruning
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• 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
• Sparse Networks from Scratch: Faster Training without Losing Performance
• Simple And Efficient Architecture Search for Convolutional Neural Networks
• Comparing Rewinding and Fine-tuning in Neural Network Pruning
• Net2Net: Accelerating Learning via Knowledge Transfer
• Picking Winning Tickets Before Training by Preserving Gradient Flow
• Stabilizing the Lottery Ticket Hypothesis
• Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterization
• Deconstructing Lottery Tickets: Zeros, Signs, and the Supermask
• Highway and Residual Networks learn Unrolled Iterative Estimation
• Dynamic Model Pruning with Feedback
• Pruning neural networks without any data by iteratively conserving synaptic flow
• The Lottery Ticket Hypothesis for Pre-trained BERT Networks
• The Lottery Tickets Hypothesis for Supervised and Self-supervised Pre-training in Computer Vision Models
• Graph Structure of Neural Networks
• Do We Actually Need Dense Over-Parameterization? In-Time Over-Parameterization in Sparse Training
• EarlyBERT: Efficient BERT Training via Early-bird Lottery Tickets
• Selfish Sparse RNN Training
• Firefly Neural Architecture Descent: a General Approach for Growing Neural Networks