Greedy Network Enlarging

• URL: http://arxiv.org/abs/2108.00177v2
• Date: Wed, 4 Aug 2021 08:07:19 GMT
• Title: Greedy Network Enlarging
• Authors: Chuanjian Liu, Kai Han, An Xiao, Yiping Deng, Wei Zhang, Chunjing Xu, Yunhe Wang
• Abstract summary: We propose a greedy network enlarging method based on the reallocation of computations. With step-by-step modifying the computations on different stages, the enlarged network will be equipped with optimal allocation and utilization of MACs. With application of our method on GhostNet, we achieve state-of-the-art 80.9% and 84.3% ImageNet top-1 accuracies.
• Score: 53.319011626986004
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent studies on deep convolutional neural networks present a simple paradigm of architecture design, i.e., models with more MACs typically achieve better accuracy, such as EfficientNet and RegNet. These works try to enlarge all the stages in the model with one unified rule by sampling and statistical methods. However, we observe that some network architectures have similar MACs and accuracies, but their allocations on computations for different stages are quite different. In this paper, we propose to enlarge the capacity of CNN models by improving their width, depth and resolution on stage level. Under the assumption that the top-performing smaller CNNs are a proper subcomponent of the top-performing larger CNNs, we propose an greedy network enlarging method based on the reallocation of computations. With step-by-step modifying the computations on different stages, the enlarged network will be equipped with optimal allocation and utilization of MACs. On EfficientNet, our method consistently outperforms the performance of the original scaling method. In particular, with application of our method on GhostNet, we achieve state-of-the-art 80.9% and 84.3% ImageNet top-1 accuracies under the setting of 600M and 4.4B MACs, respectively.

Related papers

• Principled Architecture-aware Scaling of Hyperparameters [69.98414153320894]
  Training a high-quality deep neural network requires choosing suitable hyperparameters, which is a non-trivial and expensive process. In this work, we precisely characterize the dependence of initializations and maximal learning rates on the network architecture. We demonstrate that network rankings can be easily changed by better training networks in benchmarks.
  arXiv  Detail & Related papers  (2024-02-27T11:52:49Z)
• Receptive Field Refinement for Convolutional Neural Networks Reliably Improves Predictive Performance [1.52292571922932]
  We present a new approach to receptive field analysis that can yield these types of theoretical and empirical performance gains. Our approach is able to improve ImageNet1K performance across a wide range of well-known, state-of-the-art (SOTA) model classes.
  arXiv  Detail & Related papers  (2022-11-26T05:27:44Z)
• EdgeNeXt: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applications [68.35683849098105]
  We introduce split depth-wise transpose attention (SDTA) encoder that splits input tensors into multiple channel groups. Our EdgeNeXt model with 1.3M parameters achieves 71.2% top-1 accuracy on ImageNet-1K. Our EdgeNeXt model with 5.6M parameters achieves 79.4% top-1 accuracy on ImageNet-1K.
  arXiv  Detail & Related papers  (2022-06-21T17:59:56Z)
• Towards Enabling Dynamic Convolution Neural Network Inference for Edge Intelligence [0.0]
  Recent advances in edge intelligence require CNN inference on edge network to increase throughput and reduce latency. To provide flexibility, dynamic parameter allocation to different mobile devices is required to implement either a predefined or defined on-the-fly CNN architecture. We propose a library-based approach to design scalable and dynamic distributed CNN inference on the fly.
  arXiv  Detail & Related papers  (2022-02-18T22:33:42Z)
• Multi-objective Evolutionary Approach for Efficient Kernel Size and Shape for CNN [12.697368516837718]
  State-of-the-art development in CNN topology, such as VGGNet and ResNet, have become increasingly accurate. These networks are computationally expensive involving billions of arithmetic operations and parameters. This paper considers optimising the computational resource consumption by reducing the size and number of kernels in convolutional layers.
  arXiv  Detail & Related papers  (2021-06-28T14:47:29Z)
• Container: Context Aggregation Network [83.12004501984043]
  Recent finding shows that a simple based solution without any traditional convolutional or Transformer components can produce effective visual representations. We present the model (CONText Ion NERtwok), a general-purpose building block for multi-head context aggregation. In contrast to Transformer-based methods that do not scale well to downstream tasks that rely on larger input image resolutions, our efficient network, named modellight, can be employed in object detection and instance segmentation networks.
  arXiv  Detail & Related papers  (2021-06-02T18:09:11Z)
• Convolution Neural Network Hyperparameter Optimization Using Simplified Swarm Optimization [2.322689362836168]
  Convolutional Neural Network (CNN) is widely used in computer vision. It is not easy to find a network architecture with better performance.
  arXiv  Detail & Related papers  (2021-03-06T00:23:27Z)
• Fusion of CNNs and statistical indicators to improve image classification [65.51757376525798]
  Convolutional Networks have dominated the field of computer vision for the last ten years. Main strategy to prolong this trend relies on further upscaling networks in size. We hypothesise that adding heterogeneous sources of information may be more cost-effective to a CNN than building a bigger network.
  arXiv  Detail & Related papers  (2020-12-20T23:24:31Z)
• Fully Dynamic Inference with Deep Neural Networks [19.833242253397206]
  Two compact networks, called Layer-Net (L-Net) and Channel-Net (C-Net), predict on a per-instance basis which layers or filters/channels are redundant and therefore should be skipped. On the CIFAR-10 dataset, LC-Net results in up to 11.9$times$ fewer floating-point operations (FLOPs) and up to 3.3% higher accuracy compared to other dynamic inference methods. On the ImageNet dataset, LC-Net achieves up to 1.4$times$ fewer FLOPs and up to 4.6% higher Top-1 accuracy than the other methods.
  arXiv  Detail & Related papers  (2020-07-29T23:17:48Z)
• Network Adjustment: Channel Search Guided by FLOPs Utilization Ratio [101.84651388520584]
  This paper presents a new framework named network adjustment, which considers network accuracy as a function of FLOPs. Experiments on standard image classification datasets and a wide range of base networks demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2020-04-06T15:51:00Z)

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.