SCHEME: Scalable Channel Mixer for Vision Transformers

• URL: http://arxiv.org/abs/2312.00412v3
• Date: Tue, 28 May 2024 07:13:39 GMT
• Title: SCHEME: Scalable Channel Mixer for Vision Transformers
• Authors: Deepak Sridhar, Yunsheng Li, Nuno Vasconcelos,
• Abstract summary: Vision Transformers have achieved impressive performance in many vision tasks. Much less research has been devoted to the channel mixer or feature mixing block (FFN or) We show that the dense connections can be replaced with a diagonal block structure that supports larger expansion ratios.
• Score: 52.605868919281086
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Vision Transformers have achieved impressive performance in many vision tasks. While the token mixer or attention block has been studied in great detail, much less research has been devoted to the channel mixer or feature mixing block (FFN or MLP), which accounts for a significant portion of of the model parameters and computation. In this work, we show that the dense MLP connections can be replaced with a block diagonal MLP structure that supports larger expansion ratios by splitting MLP features into groups. To improve the feature clusters formed by this structure we propose the use of a lightweight, parameter-free, channel covariance attention (CCA) mechanism as a parallel branch during training. This enables gradual feature mixing across channel groups during training whose contribution decays to zero as the training progresses to convergence. In result, the CCA block can be discarded during inference, enabling enhanced performance at no additional computational cost. The resulting $\textit{Scalable CHannEl MixEr}$ (SCHEME) can be plugged into any ViT architecture to obtain a gamut of models with different trade-offs between complexity and performance by controlling the block diagonal MLP structure. This is shown by the introduction of a new family of SCHEMEformer models. Experiments on image classification, object detection, and semantic segmentation, with different ViT backbones, consistently demonstrate substantial accuracy gains over existing designs, especially for lower complexity regimes. The SCHEMEformer family is shown to establish new Pareto frontiers for accuracy vs FLOPS, accuracy vs model size, and accuracy vs throughput, especially for fast transformers of small size.

Related papers

• PointMT: Efficient Point Cloud Analysis with Hybrid MLP-Transformer Architecture [46.266960248570086]
  This study tackles the quadratic complexity of the self-attention mechanism by introducing a complexity local attention mechanism for effective feature aggregation. We also introduce a parameter-free channel temperature adaptation mechanism that adaptively adjusts the attention weight distribution in each channel. We show that PointMT achieves performance comparable to state-of-the-art methods while maintaining an optimal balance between performance and accuracy.
  arXiv  Detail & Related papers  (2024-08-10T10:16:03Z)
• Hierarchical Associative Memory, Parallelized MLP-Mixer, and Symmetry Breaking [6.9366619419210656]
  Transformers have established themselves as the leading neural network model in natural language processing. Recent research has explored replacing attention modules with other mechanisms, including those described by MetaFormers. This paper integrates Krotov's hierarchical associative memory with MetaFormers, enabling a comprehensive representation of the Transformer block.
  arXiv  Detail & Related papers  (2024-06-18T02:42:19Z)
• CMamba: Channel Correlation Enhanced State Space Models for Multivariate Time Series Forecasting [18.50360049235537]
  Mamba, a state space model, has emerged with robust sequence and feature mixing capabilities. Capturing cross-channel dependencies is critical in enhancing performance of time series prediction. We introduce a refined Mamba variant tailored for time series forecasting.
  arXiv  Detail & Related papers  (2024-06-08T01:32:44Z)
• DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detection [52.74152717667157]
  We propose a lightweight module called Dual Attention Module (DAM) for capturing cross-dimension interaction relationships in-temporal skeletal data. It employs the frame attention mechanism to identify the most significant frames and the skeleton attention mechanism to capture broader relationships across fixed partitions with minimal parameters and flops.
  arXiv  Detail & Related papers  (2024-06-05T06:18:03Z)
• Parameter Efficient Fine-tuning via Cross Block Orchestration for Segment Anything Model [81.55141188169621]
  We equip PEFT with a cross-block orchestration mechanism to enable the adaptation of the Segment Anything Model (SAM) to various downstream scenarios. We propose an intra-block enhancement module, which introduces a linear projection head whose weights are generated from a hyper-complex layer. Our proposed approach consistently improves the segmentation performance significantly on novel scenarios with only around 1K additional parameters.
  arXiv  Detail & Related papers  (2023-11-28T11:23:34Z)
• TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting [13.410217680999459]
  Transformers have gained popularity in time series forecasting for their ability to capture long-sequence interactions. High memory and computing requirements pose a critical bottleneck for long-term forecasting. We propose TSMixer, a lightweight neural architecture composed of multi-layer perceptron (MLP) modules.
  arXiv  Detail & Related papers  (2023-06-14T06:26:23Z)
• Parameterization of Cross-Token Relations with Relative Positional Encoding for Vision MLP [52.25478388220691]
  Vision multi-layer perceptrons (MLPs) have shown promising performance in computer vision tasks. They use token-mixing layers to capture cross-token interactions, as opposed to the multi-head self-attention mechanism used by Transformers. We propose a new positional spacial gating unit (PoSGU) to efficiently encode the cross-token relations for token mixing.
  arXiv  Detail & Related papers  (2022-07-15T04:18:06Z)
• Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers [55.90468016961356]
  We propose an efficient token mixer that learns to mix in the Fourier domain. AFNO is based on a principled foundation of operator learning. It can handle a sequence size of 65k and outperforms other efficient self-attention mechanisms.
  arXiv  Detail & Related papers  (2021-11-24T05:44:31Z)
• A Battle of Network Structures: An Empirical Study of CNN, Transformer, and MLP [121.35904748477421]
  Convolutional neural networks (CNN) are the dominant deep neural network (DNN) architecture for computer vision. Transformer and multi-layer perceptron (MLP)-based models, such as Vision Transformer and Vision-Mixer, started to lead new trends. In this paper, we conduct empirical studies on these DNN structures and try to understand their respective pros and cons.
  arXiv  Detail & Related papers  (2021-08-30T06:09:02Z)
• Rethinking Token-Mixing MLP for MLP-based Vision Backbone [34.47616917228978]
  We propose an improved structure as termed Circulant Channel-Specific (CCS) token-mixing benchmark, which is spatial-invariant and channel-specific. It takes fewer parameters but achieves higher classification accuracy on ImageNet1K.
  arXiv  Detail & Related papers  (2021-06-28T17:59:57Z)

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.