Papers with Code 아카이브
Data Poisoning
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 0
- 보존된 평가 기록
- 0
Data Poisoning is an adversarial attack that tries to manipulate the training dataset in order to control the prediction behavior of a trained model such that the model will label malicious examples into a desired classes (e.g., labeling spam e-mails as safe).
Source: Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/bboylyg/backdoorllm
공식 코드BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models
https://github.com/ben-schneider-code/universal-backdoor-attacks
공식 코드https://github.com/bigml-cs-ucla/roclip
공식 코드Robust Contrastive Language-Image Pre-training against Data Poisoning and Backdoor Attacks
https://github.com/bigml-cs-ucla/safeclip
공식 코드Better Safe than Sorry: Pre-training CLIP against Targeted Data Poisoning and Backdoor Attacks
https://github.com/charlesjin/compatibility_clustering
공식 코드Incompatibility Clustering as a Defense Against Backdoor Poisoning Attacks
https://github.com/chengyi-chris/upure
공식 코드https://github.com/cristinalan/flowmur
공식 코드FlowMur: A Stealthy and Practical Audio Backdoor Attack with Limited Knowledge
https://github.com/csml-iit-ucl/bioptexps
공식 코드Bilevel Optimization with a Lower-level Contraction: Optimal Sample Complexity without Warm-start
https://github.com/dayu11/availability-attacks-create-shortcuts
공식 코드https://github.com/dessertlab/targeted-data-poisoning-attacks
공식 코드Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks
https://github.com/ebagdasa/backdoor_federated_learning
공식 코드https://github.com/eceisik/ARFED
공식 코드ARFED: Attack-Resistant Federated averaging based on outlier elimination
https://github.com/ehsannowroozi/federatedlearning_poison_lf_fp
공식 코드https://github.com/emory-aims/cae
공식 코드Classification Auto-Encoder based Detector against Diverse Data Poisoning Attacks
https://github.com/eyalperry88/lethean
공식 코드https://github.com/facebookresearch/radioactive_data
공식 코드https://github.com/foreverzyh/defend_framework
공식 코드https://github.com/ftramer/facecure
공식 코드https://github.com/git-disl/DataPoisoning_FL
공식 코드https://github.com/gsiros/drop
공식 코드DROP: Poison Dilution via Knowledge Distillation for Federated Learning
https://github.com/gvcl/iwseg-sar-poison
공식 코드Adversarial Robustness of Deep Learning Models for Inland Water Body Segmentation from SAR Images
https://github.com/h-aboutalebi/deepfakeart
공식 코드https://github.com/harpreetvirkk/beas
공식 코드BEAS: Blockchain Enabled Asynchronous & Secure Federated Machine Learning
https://github.com/hendrycks/robustness
공식 코드https://github.com/hong-xian/backdoor-attack
공식 코드Generalization Bound and New Algorithm for Clean-Label Backdoor Attack
https://github.com/hong-xian/gue
공식 코드https://github.com/intrudetection/robevalanodetect
공식 코드Robustness Evaluation of Deep Unsupervised Learning Algorithms for Intrusion Detection Systems
https://github.com/jagielski/auditing-dpsgd
공식 코드Auditing Differentially Private Machine Learning: How Private is Private SGD?
https://github.com/janzahalka/trainwreck
공식 코드Trainwreck: A damaging adversarial attack on image classifiers
https://github.com/jiamingzhang94/Unlearnable-Clusters
공식 코드Unlearnable Clusters: Towards Label-agnostic Unlearnable Examples
https://github.com/jihunhamm/bilevel-penalty
공식 코드https://github.com/jimmy-di/camouflage-poisoning
공식 코드Hidden Poison: Machine Unlearning Enables Camouflaged Poisoning Attacks
https://github.com/jinxhy/tflexattack
공식 코드https://github.com/jjy1994/BaggingCertifyDataPoisoning
공식 코드Intrinsic Certified Robustness of Bagging against Data Poisoning Attacks
https://github.com/k1rezaei/DPA-election
공식 코드Run-Off Election: Improved Provable Defense against Data Poisoning Attacks
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.