Picture for Ali Mirzaeian

Ali Mirzaeian

Adaptive-Gravity: A Defense Against Adversarial Samples

Add code
Apr 07, 2022
Figure 1 for Adaptive-Gravity: A Defense Against Adversarial Samples
Figure 2 for Adaptive-Gravity: A Defense Against Adversarial Samples
Figure 3 for Adaptive-Gravity: A Defense Against Adversarial Samples
Figure 4 for Adaptive-Gravity: A Defense Against Adversarial Samples
Viaarxiv icon

Learning Diverse Latent Representations for Improving the Resilience to Adversarial Attacks

Add code
Jul 12, 2020
Figure 1 for Learning Diverse Latent Representations for Improving the Resilience to Adversarial Attacks
Figure 2 for Learning Diverse Latent Representations for Improving the Resilience to Adversarial Attacks
Figure 3 for Learning Diverse Latent Representations for Improving the Resilience to Adversarial Attacks
Figure 4 for Learning Diverse Latent Representations for Improving the Resilience to Adversarial Attacks
Viaarxiv icon

Cluster-Based Partitioning of Convolutional Neural Networks, A Solution for Computational Energy and Complexity Reduction

Add code
Jul 12, 2020
Figure 1 for Cluster-Based Partitioning of Convolutional Neural Networks, A Solution for Computational Energy and Complexity Reduction
Figure 2 for Cluster-Based Partitioning of Convolutional Neural Networks, A Solution for Computational Energy and Complexity Reduction
Figure 3 for Cluster-Based Partitioning of Convolutional Neural Networks, A Solution for Computational Energy and Complexity Reduction
Figure 4 for Cluster-Based Partitioning of Convolutional Neural Networks, A Solution for Computational Energy and Complexity Reduction
Viaarxiv icon

Code-Bridged Classifier : A Low or Negative Overhead Defense for Making a CNN Classifier Robust Against Adversarial Attacks

Add code
Jan 16, 2020
Figure 1 for Code-Bridged Classifier : A Low or Negative Overhead Defense for Making a CNN Classifier Robust Against Adversarial Attacks
Figure 2 for Code-Bridged Classifier : A Low or Negative Overhead Defense for Making a CNN Classifier Robust Against Adversarial Attacks
Figure 3 for Code-Bridged Classifier : A Low or Negative Overhead Defense for Making a CNN Classifier Robust Against Adversarial Attacks
Figure 4 for Code-Bridged Classifier : A Low or Negative Overhead Defense for Making a CNN Classifier Robust Against Adversarial Attacks
Viaarxiv icon

TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs

Add code
Oct 14, 2019
Figure 1 for TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs
Figure 2 for TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs
Figure 3 for TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs
Figure 4 for TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs
Viaarxiv icon

NESTA: Hamming Weight Compression-Based Neural Proc. Engine

Add code
Oct 01, 2019
Figure 1 for NESTA: Hamming Weight Compression-Based Neural Proc. Engine
Figure 2 for NESTA: Hamming Weight Compression-Based Neural Proc. Engine
Figure 3 for NESTA: Hamming Weight Compression-Based Neural Proc. Engine
Figure 4 for NESTA: Hamming Weight Compression-Based Neural Proc. Engine
Viaarxiv icon