khalooei / LSA
LSA : Layer Sustainability Analysis framework for the analysis of layer vulnerability in a given neural network. LSA can be a helpful toolkit to assess deep neural networks and to extend the adversarial training approaches towards improving the sustainability of model layers via layer monitoring and analysis.
☆16Updated 2 years ago
Related projects ⓘ
Alternatives and complementary repositories for LSA
- CIFS: Improving Adversarial Robustness of CNNs via Channel-wise Importance-based Feature Selection☆20Updated 3 years ago
- ☆10Updated 2 years ago
- [ICLR 2022] Reliable Adversarial Distillation with Unreliable Teachers☆20Updated 2 years ago
- Code for Boosting fast adversarial training with learnable adversarial initialization (TIP2022)☆29Updated last year
- Implementation of ECCV 2020 "Sparse Adversarial Attack via Perturbation Factorization"☆27Updated 4 years ago
- Codes for ICCV 2021 paper "AGKD-BML: Defense Against Adversarial Attack by Attention Guided Knowledge Distillation and Bi-directional Met…☆12Updated 2 years ago
- This is the official code for "Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better"☆39Updated 3 years ago
- [NeurIPS 2021] “When does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning?”☆46Updated 3 years ago
- Membership Inference Attacks and Defenses in Neural Network Pruning☆28Updated 2 years ago
- Query-Efficient Data-Free Learning from Black-Box Models☆19Updated last year
- Enhancing Intrinsic Adversarial Robustness via Feature Pyramid Decoder(CVPR2020)☆11Updated 4 years ago
- Data-Efficient Backdoor Attacks☆18Updated 2 years ago
- Code for our NeurIPS 2020 paper Backpropagating Linearly Improves Transferability of Adversarial Examples.☆42Updated last year
- A Unified Approach to Interpreting and Boosting Adversarial Transferability (ICLR2021)☆28Updated 2 years ago
- Implementation of adversarial training under fast-gradient sign method (FGSM), projected gradient descent (PGD) and CW using Wide-ResNet-…☆38Updated 4 years ago
- [NeurIPS 2020] code for "Boundary thickness and robustness in learning models"☆18Updated 3 years ago
- source code for NeurIPS21 paper robabilistic Margins for Instance Reweighting in Adversarial Training☆10Updated 2 years ago
- ☆17Updated last year
- Implementation for <Robust Weight Perturbation for Adversarial Training> in IJCAI'22.☆14Updated 2 years ago
- the paper "Geometry-aware Instance-reweighted Adversarial Training" ICLR 2021 oral☆57Updated 3 years ago
- [ICLR 2021] "Robust Overfitting may be mitigated by properly learned smoothening" by Tianlong Chen*, Zhenyu Zhang*, Sijia Liu, Shiyu Chan…☆46Updated 2 years ago
- [NeurIPS 2021] "Class-Disentanglement and Applications in Adversarial Detection and Defense"☆43Updated 2 years ago
- Towards Efficient and Effective Adversarial Training, NeurIPS 2021☆16Updated 2 years ago
- PyTorch implementation of BPDA+EOT attack to evaluate adversarial defense with an EBM☆23Updated 4 years ago
- ATTA (Efficient Adversarial Training with Transferable Adversarial Examples)☆32Updated 4 years ago
- Source code for ECCV 2022 Poster: Data-free Backdoor Removal based on Channel Lipschitzness☆29Updated last year
- Fighting Gradients with Gradients: Dynamic Defenses against Adversarial Attacks☆37Updated 3 years ago
- [ICLR 2023, Spotlight] Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning☆28Updated 11 months ago
- The source code for ICML2021 paper When Does Data Augmentation Help With Membership Inference Attacks?☆8Updated 3 years ago
- ☆55Updated 2 years ago