Achieving adversarial robustness via sparsity

Liao, Ningyi; Wang, Shufan; Xiang, Liyao; Ye, Nanyang; Shao, Shuo; Chu, Pengzhi

doi:10.1007/s10994-021-06049-9

Cited by 6 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Soft threshold weight reparameterization (STR) [22] smoothly induces the sparsity while learning pruning thresholds. The authors of [23] proposed a novel adversarial training method called inverse weight inheritance, which imposed sparse weight distribution on a large network by inheriting weights from a small network, thereby improving the robustness of the large network. However, these methods are not able to realize an extremely large parameter compression ratio.…”

Section: Related Workmentioning

confidence: 99%

Towards Super Compressed Neural Networks for Object Identification: Quantized Low-Rank Tensor Decomposition with Self-Attention

Liu,

Wang,

et al. 2024

Electronics

View full text Add to dashboard Cite

Deep convolutional neural networks have a large number of parameters and require a significant number of floating-point operations during computation, which limits their deployment in situations where the storage space is limited and computational resources are insufficient, such as in mobile phones and small robots. Many network compression methods have been proposed to address the aforementioned issues, including pruning, low-rank decomposition, quantization, etc. However, these methods typically fail to achieve a significant compression ratio in terms of the parameter count. Even when high compression rates are achieved, the network’s performance is often significantly deteriorated, making it difficult to perform tasks effectively. In this study, we propose a more compact representation for neural networks, named Quantized Low-Rank Tensor Decomposition (QLTD), to super compress deep convolutional neural networks. Firstly, we employed low-rank Tucker decomposition to compress the pre-trained weights. Subsequently, to further exploit redundancies within the core tensor and factor matrices obtained through Tucker decomposition, we employed vector quantization to partition and cluster the weights. Simultaneously, we introduced a self-attention module for each core tensor and factor matrix to enhance the training responsiveness in critical regions. The object identification results in the CIFAR10 experiment showed that QLTD achieved a compression ratio of 35.43×, with less than 1% loss in accuracy and a compression ratio of 90.61×, with less than a 2% loss in accuracy. QLTD was able to achieve a significant compression ratio in terms of the parameter count and realize a good balance between compressing parameters and maintaining identification accuracy.

show abstract

Section: Related Workmentioning

confidence: 99%