Pruning Self-attentions into Convolutional Layers in Single Path

• URL: http://arxiv.org/abs/2111.11802v4
• Date: Tue, 16 Jan 2024 09:18:27 GMT
• Title: Pruning Self-attentions into Convolutional Layers in Single Path
• Authors: Haoyu He, Jianfei Cai, Jing Liu, Zizheng Pan, Jing Zhang, Dacheng Tao, Bohan Zhuang
• Abstract summary: Vision Transformers (ViTs) have achieved impressive performance over various computer vision tasks. We propose Single-Path Vision Transformer pruning (SPViT) to efficiently and automatically compress the pre-trained ViTs. Our SPViT can trim 52.0% FLOPs for DeiT-B and get an impressive 0.6% top-1 accuracy gain simultaneously.
• Score: 89.55361659622305
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision Transformers (ViTs) have achieved impressive performance over various computer vision tasks. However, modeling global correlations with multi-head self-attention (MSA) layers leads to two widely recognized issues: the massive computational resource consumption and the lack of intrinsic inductive bias for modeling local visual patterns. To solve both issues, we devise a simple yet effective method named Single-Path Vision Transformer pruning (SPViT), to efficiently and automatically compress the pre-trained ViTs into compact models with proper locality added. Specifically, we first propose a novel weight-sharing scheme between MSA and convolutional operations, delivering a single-path space to encode all candidate operations. In this way, we cast the operation search problem as finding which subset of parameters to use in each MSA layer, which significantly reduces the computational cost and optimization difficulty, and the convolution kernels can be well initialized using pre-trained MSA parameters. Relying on the single-path space, we introduce learnable binary gates to encode the operation choices in MSA layers. Similarly, we further employ learnable gates to encode the fine-grained MLP expansion ratios of FFN layers. In this way, our SPViT optimizes the learnable gates to automatically explore from a vast and unified search space and flexibly adjust the MSA-FFN pruning proportions for each individual dense model. We conduct extensive experiments on two representative ViTs showing that our SPViT achieves a new SOTA for pruning on ImageNet-1k. For example, our SPViT can trim 52.0% FLOPs for DeiT-B and get an impressive 0.6% top-1 accuracy gain simultaneously. The source code is available at https://github.com/ziplab/SPViT.

Related papers

• Sparse-Tuning: Adapting Vision Transformers with Efficient Fine-tuning and Inference [14.030836300221756]
  textbfSparse-Tuning is a novel PEFT method that accounts for the information redundancy in images and videos. Sparse-Tuning minimizes the quantity of tokens processed at each layer, leading to a quadratic reduction in computational and memory overhead. Our results show that our Sparse-Tuning reduces GFLOPs to textbf62%-70% of the original ViT-B while achieving state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-05-23T15:34:53Z)
• Edge-MoE: Memory-Efficient Multi-Task Vision Transformer Architecture with Task-level Sparsity via Mixture-of-Experts [60.1586169973792]
  M$3$ViT is the latest multi-task ViT model that introduces mixture-of-experts (MoE) MoE achieves better accuracy and over 80% reduction computation but leaves challenges for efficient deployment on FPGA. Our work, dubbed Edge-MoE, solves the challenges to introduce the first end-to-end FPGA accelerator for multi-task ViT with a collection of architectural innovations.
  arXiv  Detail & Related papers  (2023-05-30T02:24:03Z)
• Parameter-efficient Tuning of Large-scale Multimodal Foundation Model [68.24510810095802]
  We propose A graceful prompt framework for cross-modal transfer (Aurora) to overcome these challenges. Considering the redundancy in existing architectures, we first utilize the mode approximation to generate 0.1M trainable parameters to implement the multimodal prompt tuning. A thorough evaluation on six cross-modal benchmarks shows that it not only outperforms the state-of-the-art but even outperforms the full fine-tuning approach.
  arXiv  Detail & Related papers  (2023-05-15T06:40:56Z)
• DilateFormer: Multi-Scale Dilated Transformer for Visual Recognition [62.95223898214866]
  We explore effective Vision Transformers to pursue a preferable trade-off between the computational complexity and size of the attended receptive field. With a pyramid architecture, we construct a Multi-Scale Dilated Transformer (DilateFormer) by stacking MSDA blocks at low-level stages and global multi-head self-attention blocks at high-level stages. Our experiment results show that our DilateFormer achieves state-of-the-art performance on various vision tasks.
  arXiv  Detail & Related papers  (2023-02-03T14:59:31Z)
• Rethinking Hierarchicies in Pre-trained Plain Vision Transformer [76.35955924137986]
  Self-supervised pre-training vision transformer (ViT) via masked image modeling (MIM) has been proven very effective. customized algorithms should be carefully designed for the hierarchical ViTs, e.g., GreenMIM, instead of using the vanilla and simple MAE for the plain ViT. This paper proposes a novel idea of disentangling the hierarchical architecture design from the self-supervised pre-training.
  arXiv  Detail & Related papers  (2022-11-03T13:19:23Z)
• ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyond [76.35955924137986]
  We propose a Vision Transformer Advanced by Exploring intrinsic IB from convolutions, i.e., ViTAE. ViTAE has several spatial pyramid reduction modules to downsample and embed the input image into tokens with rich multi-scale context. We obtain the state-of-the-art classification performance, i.e., 88.5% Top-1 classification accuracy on ImageNet validation set and the best 91.2% Top-1 accuracy on ImageNet real validation set.
  arXiv  Detail & Related papers  (2022-02-21T10:40:05Z)
• SimViT: Exploring a Simple Vision Transformer with sliding windows [3.3107339588116123]
  We introduce a vision Transformer named SimViT, to incorporate spatial structure and local information into the vision Transformers. SimViT extracts multi-scale hierarchical features from different layers for dense prediction tasks. Our SimViT-Micro only needs 3.3M parameters to achieve 71.1% top-1 accuracy on ImageNet-1k dataset.
  arXiv  Detail & Related papers  (2021-12-24T15:18:20Z)
• FQ-ViT: Fully Quantized Vision Transformer without Retraining [13.82845665713633]
  We present a systematic method to reduce the performance degradation and inference complexity of Quantized Transformers. We are the first to achieve comparable accuracy degradation (1%) on fully quantized Vision Transformers.
  arXiv  Detail & Related papers  (2021-11-27T06:20:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.