paper-with-me

Papers

An Entropy-based Pruning Method for CNN Compression

2017-06-19 · Jian-Hao Luo, Jianxin Wu

This paper aims to simultaneously accelerate and compress off-the-shelf CNN models via filter pruning strategy. The importance of each filter is evaluated by the proposed entropy-based method first. Then several unimportant filters are discarded to get a smaller CNN model. Finally, fine-tuning is adopted to recover its generalization ability which is damaged during filter pruning. Our method can reduce the size of intermediate activations, which would dominate most memory footprint during model training stage but is less concerned in previous compression methods. Experiments on the ILSVRC-12 benchmark demonstrate the effectiveness of our method. Compared with previous filter importance evaluation criteria, our entropy-based method obtains better performance. We achieve 3.3x speed-up and 16.64x compression on VGG-16, 1.54x acceleration and 1.47x compression on ResNet-50, both with about 1% top-5 accuracy decrease.

📄 PDF Abstract BibTeX arXiv:1706.05791

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

Encoding Weights of Irregular Sparsity for Fixed-to-Fixed Model Compression

2021-05-05 · ICLR 2022 4 · Baeseong Park, Se Jung Kwon, Daehwan Oh, Byeongwook Kim 외

Even though fine-grained pruning techniques achieve a high compression ratio, conventional sparsity representations (such as CSR) associated with irregular sparsity degrade parallelism significantly. Practical pruning me…

Model Compression

Entropy-Constrained Training of Deep Neural Networks

2018-12-18 · Simon Wiedemann, Arturo Marban, Klaus-Robert Müller, Wojciech Samek

We propose a general framework for neural network compression that is motivated by the Minimum Description Length (MDL) principle. For that we first derive an expression for the entropy of a neural network, which measure…

Neural Network Compression

Entropy Meets Importance: A Unified Head Importance-Entropy Score for Stable and Efficient Transformer Pruning

2025-10-10 · Minsik Choi, Hyegang Son, Changhoon Kim, Young Geun Kim arxiv

Transformer-based models have achieved remarkable performance in NLP tasks. However, their structural characteristics-multiple layers and attention heads-introduce efficiency challenges in inference and deployment. To ad…

Model Compression

2PFPCE: Two-Phase Filter Pruning Based on Conditional Entropy

2018-09-06 · Chuhan Min, Aosen Wang, Yiran Chen, Wenyao Xu 외

Deep Convolutional Neural Networks~(CNNs) offer remarkable performance of classifications and regressions in many high-dimensional problems and have been widely utilized in real-word cognitive applications. However, high…

Edge-computingNeural Network CompressionVocal Bursts Valence Prediction

Can Unstructured Pruning Reduce the Depth in Deep Neural Networks?

2023-08-12 · Zhu Liao, Victor Quétu, Van-Tam Nguyen, Enzo Tartaglione

Pruning is a widely used technique for reducing the size of deep neural networks while maintaining their performance. However, such a technique, despite being able to massively compress deep models, is hardly able to rem…