Msranet: Learning Discriminative Embeddings for Speaker Verification Via Channel and Spatial Attention Mechanism in Alterable Scenarios

Zheng, Qiuyu; Chen, Zengzhao; Liu, Hai; Lu, Yuanyuan; Li, Jiawen

doi:10.2139/ssrn.4178119

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The evolution after the AlexNet can be mainly summarized in two ways: sequence-focused models by increasing the depth of networks and nonsequence-focused models by adding units or modules. Sequence focused models: by leveraging the sequence information through layers to improve the model performance, such as VGG16 and VGG19 [9], MSRANet [10]. All of these models are focused on using multiple convolutional layers and activation layers to increase the depth of network, which can extract deeper and better features than the simple structure.…”

Section: Convolutional Neural Network Structuresmentioning

confidence: 99%

Hybrid CNN -Interpreter: Interpret local and global contexts for CNN-based Models

Yang¹,

Guan²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Convolutional neural network (CNN) models have seen advanced improvements in performance in various domains, but lack of interpretability is a major barrier to assurance and regulation during operation for acceptance and deployment of AI-assisted applications. There have been many works on input interpretability focusing on analyzing the input-output relations, but the internal logic of models has not been clarified in the current mainstream interpretability methods. In this study, we propose a novel hybrid CNN-interpreter through: (1) An original forward propagation mechanism to examine the layer-specific prediction results for local interpretability. (2) A new global interpretability that indicates the feature correlation and filter importance effects. By combining the local and global interpretabilities, hybrid CNN-interpreter enables us to have a solid understanding and monitoring of model context during the whole learning process with detailed and consistent representations. Finally, the proposed interpretabilities have been demonstrated to adapt to various CNN-based model structures.

show abstract

Section: Convolutional Neural Network Structuresmentioning

confidence: 99%