Redirigiendo al acceso original de articulo en 16 segundos...
Inicio  /  Informatics  /  Vol: 10 Par: 3 (2023)  /  Artículo
ARTÍCULO
TITULO

Sp2PS: Pruning Score by Spectral and Spatial Evaluation of CAM Images

Diego Renza and Dora Ballesteros    

Resumen

CNN models can have millions of parameters, which makes them unattractive for some applications that require fast inference times or small memory footprints. To overcome this problem, one alternative is to identify and remove weights that have a small impact on the loss function of the algorithm, which is known as pruning. Typically, pruning methods are compared in terms of performance (e.g., accuracy), model size and inference speed. However, it is unusual to evaluate whether a pruned model preserves regions of importance in an image when performing inference. Consequently, we propose a metric to assess the impact of a pruning method based on images obtained by model interpretation (specifically, class activation maps). These images are spatially and spectrally compared and integrated by the harmonic mean for all samples in the test dataset. The results show that although the accuracy in a pruned model may remain relatively constant, the areas of attention for decision making are not necessarily preserved. Furthermore, the performance of pruning methods can be easily compared as a function of the proposed metric.

 Artículos similares

       
 
Leila Ben Letaifa and Jean-Luc Rouas    
Transformer models are being increasingly used in end-to-end speech recognition systems for their performance. However, their substantial size poses challenges for deploying them in real-world applications. These models heavily rely on attention and feed... ver más
Revista: Algorithms

 
Sichao Zhuo, Xiaoming Zhang, Ziyi Chen, Wei Wei, Fang Wang, Quanlong Li and Yufan Guan    
With the development of Industry 4.0, although some smart meters have appeared on the market, traditional mechanical meters are still widely used due to their long-standing presence and the difficulty of modifying or replacing them in large quantities. M... ver más
Revista: Applied Sciences

 
Zhuo Li, Hengyi Li and Lin Meng    
Currently, with the rapid development of deep learning, deep neural networks (DNNs) have been widely applied in various computer vision tasks. However, in the pursuit of performance, advanced DNN models have become more complex, which has led to a large ... ver más
Revista: Computers

 
Sicong Liu, Qingcheng Fan, Chunjiang Zhao and Shuqin Li    
Animal resources are significant to human survival and development and the ecosystem balance. Automated multi-animal object detection is critical in animal research and conservation and ecosystem monitoring. The objective is to design a model that mitiga... ver más
Revista: Information

 
Chun-Myoung Noh, Jun-Gyo Jang, Sung-Soo Kim, Soon-Sup Lee, Sung-Chul Shin and Jae-Chul Lee    
With increasing interest in smart factories, considerable attention has been paid to the development of deep-learning-based quality inspection systems. Deep-learning-based quality inspection helps productivity improvements by solving the limitations of e... ver más
Revista: Applied Sciences