How Well Do Sparse Imagenet Models Transfer?
This addresses the problem of model efficiency for practitioners in computer vision by showing that pruning does not necessarily harm transfer learning, potentially enabling faster and more resource-efficient deployment.
The study investigated whether sparse convolutional neural networks (CNNs) pruned from ImageNet can transfer effectively to downstream tasks, finding that they can match or outperform dense models in transfer accuracy even at high sparsity levels, with up to 80% sparsity showing comparable performance and leading to inference and training speedups.
Transfer learning is a classic paradigm by which models pretrained on large "upstream" datasets are adapted to yield good results on "downstream" specialized datasets. Generally, more accurate models on the "upstream" dataset tend to provide better transfer accuracy "downstream". In this work, we perform an in-depth investigation of this phenomenon in the context of convolutional neural networks (CNNs) trained on the ImageNet dataset, which have been pruned - that is, compressed by sparsifying their connections. We consider transfer using unstructured pruned models obtained by applying several state-of-the-art pruning methods, including magnitude-based, second-order, re-growth, lottery-ticket, and regularization approaches, in the context of twelve standard transfer tasks. In a nutshell, our study shows that sparse models can match or even outperform the transfer performance of dense models, even at high sparsities, and, while doing so, can lead to significant inference and even training speedups. At the same time, we observe and analyze significant differences in the behaviour of different pruning methods.