paper-with-me

홈 › Papers

Vision Transformer Pruning

2021-04-17 · Mingjian Zhu, Yehui Tang, Kai Han

Vision transformer has achieved competitive performance on a variety of computer vision applications. However, their storage, run-time memory, and computational demands are hindering the deployment to mobile devices. Here we present a vision transformer pruning approach, which identifies the impacts of dimensions in each layer of transformer and then executes pruning accordingly. By encouraging dimension-wise sparsity in the transformer, important dimensions automatically emerge. A great number of dimensions with small importance scores can be discarded to achieve a high pruning ratio without significantly compromising accuracy. The pipeline for vision transformer pruning is as follows: 1) training with sparsity regularization; 2) pruning dimensions of linear projections; 3) fine-tuning. The reduced parameters and FLOPs ratios of the proposed algorithm are well evaluated and analyzed on ImageNet dataset to demonstrate the effectiveness of our proposed method.

📄 PDF Abstract BibTeX arXiv:2104.08500

Code (2)

cydia2018/vit-cifar10-pruning pytorch
kentaroy47/vision-transformers-cifar10 pytorch

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Pruning 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Vision Transformer The Vision Transformer, or ViT, is a model for image classification that employs a Transformer-like architecture over…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Searching Intrinsic Dimensions of Vision Transformers

2022-04-16 · Fanghui Xue, Biao Yang, Yingyong Qi, Jack Xin

It has been shown by many researchers that transformers perform as well as convolutional neural networks in many computer vision tasks. Meanwhile, the large computational costs of its attention module hinder further stud…

image-classificationImage Classificationobject-detectionObject Detection

UPDP: A Unified Progressive Depth Pruner for CNN and Vision Transformer

2024-01-12 · Ji Liu, Dehua Tang, Yuanxian Huang, Li Zhang 외

Traditional channel-wise pruning methods by reducing network channels struggle to effectively prune efficient CNN models with depth-wise convolutional layers and certain efficient modules, such as popular inverted residu…

Sparse then Prune: Toward Efficient Vision Transformers

2023-07-22 · Yogi Prasetyo, Novanto Yudistira, Agus Wahyu Widodo

The Vision Transformer architecture is a deep learning model inspired by the success of the Transformer model in Natural Language Processing. However, the self-attention mechanism, large number of parameters, and the req…

image-classificationImage Classification

PQV-Mobile: A Combined Pruning and Quantization Toolkit to Optimize Vision Transformers for Mobile Applications

2024-08-15 · Kshitij Bhardwaj

While Vision Transformers (ViTs) are extremely effective at computer vision tasks and are replacing convolutional neural networks as the new state-of-the-art, they are complex and memory-intensive models. In order to eff…

Quantization

Entropy Reveals Block Importance in Masked Self-Supervised Vision Transformers

2026-02-03 · Peihao Xiang, Kaida Wu, Ou Bai arxiv

Masked self-supervised vision transformers have become a dominant pretraining paradigm, yet their substantial model size poses significant challenges for resource-constrained deployment and efficient transfer learning. A…

Transfer LearningModel Compression