Learning Instance-wise Sparsity for Accelerating Deep Models

被引:0
|
作者
Liu, Chuanjian [1 ]
Wang, Yunhe [1 ]
Han, Kai [1 ]
Xu, Chunjing [1 ]
Xu, Chang [2 ]
机构
[1] Huawei Noahs Ark Lab, Beijing, Peoples R China
[2] Univ Sydney, FEIT, Sch Comp Sci, Sydney, NSW, Australia
基金
澳大利亚研究理事会;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Exploring deep convolutional neural networks of high efficiency and low memory usage is very essential for a wide variety of machine learning tasks. Most of existing approaches used to accelerate deep models by manipulating parameters or filters without data, e.g., pruning and decomposition. In contrast, we study this problem from a different perspective by respecting the difference between data. An instance-wise feature pruning is developed by identifying informative features for different instances. Specifically, by investigating a feature decay regularization, we expect intermediate feature maps of each instance in deep neural networks to be sparse while preserving the overall network performance. During online inference, subtle features of input images extracted by intermediate layers of a well-trained neural network can be eliminated to accelerate the subsequent calculations. We further take coefficient of variation as a measure to select the layers that are appropriate for acceleration. Extensive experiments conducted on benchmark datasets and networks demonstrate the effectiveness of the proposed method.
引用
收藏
页码:3001 / 3007
页数:7
相关论文
共 50 条
  • [31] Instance-wise points-to analysis for loop-based dependence testing
    Wu, Peng
    Feautrier, Paul
    Padua, David
    Sura, Zehra
    Proceedings of the International Conference on Supercomputing, 2002, : 262 - 273
  • [32] SparseRT: Accelerating Unstructured Sparsity on GPUs for Deep Learning Inference
    Wang, Ziheng
    PACT '20: PROCEEDINGS OF THE ACM INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES, 2020, : 31 - 42
  • [33] OPTIMUM FEATURE ORDERING FOR DYNAMIC INSTANCE-WISE JOINT FEATURE SELECTION AND CLASSIFICATION
    Liyanage, Yasitha Warahena
    Zois, Daphney-Stavroula
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 3370 - 3374
  • [34] Asymptotically Unbiased Instance-wise Regularized Partial AUC Optimization: Theory and Algorithm
    Shao, Huiyang
    Xu, Qianqian
    Yang, Zhiyong
    Bao, Shilong
    Huang, Qingming
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [35] Objformer: Boosting 3D object detection via instance-wise interaction
    Tao, Manli
    Zhao, Chaoyang
    Tang, Ming
    Wang, Jinqiao
    PATTERN RECOGNITION, 2024, 146
  • [36] What went wrong and when? Instance-wise feature importance for time-series black-box models
    Tonekaboni, Sana
    Joshi, Shalmali
    Campbell, Kieran R.
    Duvenaud, David
    Goldenberg, Anna
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [37] Neural Network Surgery: Injecting Data Patterns into Pre-trained Models with Minimal Instance-wise Side Effects
    Zhang, Zhiyuan
    Ren, Xuancheng
    Su, Qi
    Sun, Xu
    He, Bin
    2021 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL-HLT 2021), 2021, : 5453 - 5466
  • [38] Enhancing accuracy of diabetes diagnosis system using instance-wise feature importance aware imputer and tabnet deep neural classifier
    Hameed, Wafaa Mustafa
    Ali, Nzar Abdulqadir
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 270
  • [39] Benchmarking Deep Learning Models for Instance Segmentation
    Jung, Sunguk
    Heo, Hyeonbeom
    Park, Sangheon
    Jung, Sung-Uk
    Lee, Kyungjae
    APPLIED SCIENCES-BASEL, 2022, 12 (17):
  • [40] X-DETR: A Versatile Architecture for Instance-wise Vision-Language Tasks
    Cai, Zhaowei
    Kwon, Gukyeong
    Ravichandran, Avinash
    Bas, Erhan
    Tu, Zhuowen
    Bhotika, Rahul
    Soatto, Stefano
    COMPUTER VISION, ECCV 2022, PT XXXVI, 2022, 13696 : 290 - 308