Open AccessPosted Content
To prune, or not to prune: exploring the efficacy of pruning for model compression
Michael H. Zhu,Suyog Gupta +1 more
Reads0
Chats0
TLDR
In this article, the authors investigate two distinct paths for model compression within the context of energy-efficient inference in resource-constrained environments and propose a new gradual pruning technique that is simple and straightforward to apply across a variety of models/datasets with minimal tuning.Abstract:Â
Model pruning seeks to induce sparsity in a deep neural network's various connection matrices, thereby reducing the number of nonzero-valued parameters in the model. Recent reports (Han et al., 2015; Narang et al., 2017) prune deep networks at the cost of only a marginal loss in accuracy and achieve a sizable reduction in model size. This hints at the possibility that the baseline models in these experiments are perhaps severely over-parameterized at the outset and a viable alternative for model compression might be to simply reduce the number of hidden units while maintaining the model's dense connection structure, exposing a similar trade-off in model size and accuracy. We investigate these two distinct paths for model compression within the context of energy-efficient inference in resource-constrained environments and propose a new gradual pruning technique that is simple and straightforward to apply across a variety of models/datasets with minimal tuning and can be seamlessly incorporated within the training process. We compare the accuracy of large, but pruned models (large-sparse) and their smaller, but dense (small-dense) counterparts with identical memory footprint. Across a broad range of neural network architectures (deep CNNs, stacked LSTM, and seq2seq LSTM models), we find large-sparse models to consistently outperform small-dense models and achieve up to 10x reduction in number of non-zero parameters with minimal loss in accuracy.read more
Citations
More filters
Proceedings ArticleDOI
Structured and tiled-based pruning of Deep Learning models targeting FPGA implementations
TL;DR: In this paper , the authors leverage the use of the Gumbel-Softmax relaxation sampling to structurally prune tiles, which benefits further hardware implementations, and additionally allows to jointly optimize with quantization.
Journal ArticleDOI
Towards Automatic Model Compression via a Unified Two-Stage Framework
TL;DR: In this paper , a unified two-stage framework for automatic model compression is proposed, which improves the optimization from two aspects: first, to predict the performance of each compression policy, and second, to search for the compression ratio allocation, the proposed Hessian matrix approximation and Knapsack problem reformulation.
Data Level Lottery Ticket Hypothesis for Vision Transformers
Xuan Shen,Zhenglun Kong,Mingfu Qin,Peiyan Dong,Geng Yuan,Xin Meng,Hao Tang,Xiaolong Ma,Yanzhi Wang +8 more
TL;DR: In this article , the authors generalized the lottery ticket hypothesis to input data consisting of image patches inspired by the input dependence of vision transformers (ViTs) and showed that there exists a subset of input image patches such that a ViT can be trained from scratch by using only this subset of patches and achieve similar accuracy to the ViTs trained by using all image patches.
Posted Content
EDCompress: Energy-Aware Model Compression with Dataflow.
TL;DR: EDCompress is proposed, an Energy-aware model compression method for various Dataflows that can effectively reduce the energy consumption of various edge devices, with different dataflow types, and find the optimal dataflow type for specific neural networks in terms of energy consumption.
Proceedings ArticleDOI
Fine-grained analysis of the transformer model for efficient pruning
Leila Ben Letaifa,Jean-Luc Rouas +1 more
TL;DR: In this paper , a fine-grained analysis of the transformer model layers in order to determine the most efficient pruning approach is presented. But it is more appropriate to prune some layers than others and underlines the importance of knowing the behavior of the layers to choose the pruning method.
References
More filters
Posted Content
Rethinking the Inception Architecture for Computer Vision
TL;DR: This work is exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization.
Posted Content
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Andrew Howard,Menglong Zhu,Bo Chen,Dmitry Kalenichenko,Weijun Wang,Tobias Weyand,M. Andreetto,Hartwig Adam +7 more
TL;DR: This work introduces two simple global hyper-parameters that efficiently trade off between latency and accuracy and demonstrates the effectiveness of MobileNets across a wide range of applications and use cases including object detection, finegrain classification, face attributes and large scale geo-localization.
Proceedings Article
Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
TL;DR: Deep Compression as mentioned in this paper proposes a three-stage pipeline: pruning, quantization, and Huffman coding to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy.
Posted Content
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Yonghui Wu,Mike Schuster,Zhifeng Chen,Quoc V. Le,Mohammad Norouzi,Wolfgang Macherey,Maxim Krikun,Yuan Cao,Qin Gao,Klaus Macherey,Jeff Klingner,Apurva Shah,Melvin Johnson,Xiaobing Liu,Ćukasz Kaiser,Stephan Gouws,Yoshikiyo Kato,Taku Kudo,Hideto Kazawa,Keith Stevens,George Kurian,Nishant Patil,Wei Wang,Cliff Young,Jason A. Smith,Jason Riesa,Alex Rudnick,Oriol Vinyals,Greg S. Corrado,Macduff Hughes,Jeffrey Dean +30 more
TL;DR: GNMT, Google's Neural Machine Translation system, is presented, which attempts to address many of the weaknesses of conventional phrase-based translation systems and provides a good balance between the flexibility of "character"-delimited models and the efficiency of "word"-delicited models.
Proceedings Article
Learning both weights and connections for efficient neural networks
TL;DR: In this paper, the authors proposed a method to reduce the storage and computation required by neural networks by an order of magnitude without affecting their accuracy by learning only the important connections using a three-step method.