reds-lab / Universal_Pert_Cert
This repo is the official implementation of the ICLR'23 paper "Towards Robustness Certification Against Universal Perturbations." We calculate the certified robustness against universal perturbations (UAP/ Backdoor) given a trained model.
☆12Updated 2 years ago
Alternatives and similar repositories for Universal_Pert_Cert:
Users that are interested in Universal_Pert_Cert are comparing it to the libraries listed below
- ☆11Updated 2 years ago
- Boosting the Transferability of Adversarial Attacks with Reverse Adversarial Perturbation (NeurIPS 2022)☆33Updated 2 years ago
- Backdoor Safety Tuning (NeurIPS 2023 & 2024 Spotlight)☆25Updated 4 months ago
- The official implementation of USENIX Security'23 paper "Meta-Sift" -- Ten minutes or less to find a 1000-size or larger clean subset on …☆18Updated last year
- Github repo for One-shot Neural Backdoor Erasing via Adversarial Weight Masking (NeurIPS 2022)☆15Updated 2 years ago
- Identification of the Adversary from a Single Adversarial Example (ICML 2023)☆9Updated 8 months ago
- Code for the paper "Deep Partition Aggregation: Provable Defenses against General Poisoning Attacks"☆12Updated 2 years ago
- Codes for NeurIPS 2021 paper "Adversarial Neuron Pruning Purifies Backdoored Deep Models"☆57Updated last year
- ☆19Updated 2 years ago
- RAB: Provable Robustness Against Backdoor Attacks☆39Updated last year
- Official Implementation of ICLR 2022 paper, ``Adversarial Unlearning of Backdoors via Implicit Hypergradient''☆54Updated 2 years ago
- Reconstructive Neuron Pruning for Backdoor Defense (ICML 2023)☆35Updated last year
- ☆19Updated 3 years ago
- ☆10Updated 4 years ago
- This repository is the official implementation of the paper "ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning…☆17Updated last year
- ☆21Updated 3 years ago
- [ICLR 2023, Spotlight] Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning☆30Updated last year
- ☆19Updated 6 months ago
- [NeurIPS 2021] Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training☆31Updated 3 years ago
- The implementation of our paper: Composite Adversarial Attacks (AAAI2021)☆30Updated 3 years ago
- Code of On L-p Robustness of Decision Stumps and Trees, ICML 2020☆10Updated 4 years ago
- Code for identifying natural backdoors in existing image datasets.☆15Updated 2 years ago
- Camouflage poisoning via machine unlearning☆17Updated 2 years ago
- [ICLR2023] Distilling Cognitive Backdoor Patterns within an Image☆34Updated 5 months ago
- ☆13Updated 3 years ago
- ☆53Updated last year
- ICCV 2021, We find most existing triggers of backdoor attacks in deep learning contain severe artifacts in the frequency domain. This Rep…☆43Updated 2 years ago
- Source code for ECCV 2022 Poster: Data-free Backdoor Removal based on Channel Lipschitzness☆30Updated 2 years ago
- [ICLR'21] Dataset Inference for Ownership Resolution in Machine Learning☆32Updated 2 years ago
- ☆11Updated 2 years ago