fedlab

[fedlab]-blog->github

benchmarks

数据划分，整合

fcube

专栏

https://www.bilibili.com/video/BV1pG4y1t7t5/?spm_id_from=333.999.list.card_archive.click&vd_source=ad42090d7d6fcdfc144126ae0e2884ac

–

侧重点在 adaptive semi

https://ijcai2022.scimeeting.cn/cn/web/program/15678

https://ijcaipcfl2022.scimeeting.cn/cn/web/index/15792_1264754__

Adaptive:

之前一篇待看Adaptive Federated Learning in Resource Constrained Edge Computing Systems

Source-Adaptive Discriminative Kernels based Network for Remote Sensing Pansharpening

Adaptive Convolutional Dictionary Network for CT Metal Artifact Reduction

ER-SAN: Enhanced-Adaptive Relation Self-Attention Network for Image Captioning

Constrained Adaptive Projection with Pretrained Features for Anomaly Detection

https://ai-paper-collector.vercel.app/r?query=Adaptive+Federated&year=&sp_year=&sp_author=&confs=AAAI&confs=ACL&confs=AISTATS&confs=BMVC&confs=CIKM&confs=COLING&confs=COLT&confs=CVPR&confs=ECCV&confs=ECIR&confs=EMNLP&confs=ICASSP&confs=ICCV&confs=ICDM&confs=ICLR&confs=ICME&confs=ICML&confs=IJCAI&confs=INTERSPEECH&confs=ISWC&confs=JMLR&confs=KDD&confs=MICCAI&confs=MLSYS&confs=MM&confs=NAACL&confs=NIPS&confs=RECSYS&confs=SIGIR&confs=TASLP&confs=TIP&confs=TKDE&confs=TOIS&confs=TPAMI&confs=VLDB&confs=WACV&confs=WSDM&confs=WWW

semi:

Region-level Contrastive and Consistency Learning for Semi-Supervised Semantic Segmentation

Uncertainty-Guided Pixel Contrastive Learning for Semi-Supervised Medical Image

https://ai-paper-collector.vercel.app/r?query=Semi-Supervised+Federated&year=&sp_year=&sp_author=&confs=AAAI&confs=ACL&confs=AISTATS&confs=BMVC&confs=CIKM&confs=COLING&confs=COLT&confs=CVPR&confs=ECCV&confs=ECIR&confs=EMNLP&confs=ICASSP&confs=ICCV&confs=ICDM&confs=ICLR&confs=ICME&confs=ICML&confs=IJCAI&confs=INTERSPEECH&confs=ISWC&confs=JMLR&confs=KDD&confs=MICCAI&confs=MLSYS&confs=MM&confs=NAACL&confs=NIPS&confs=RECSYS&confs=SIGIR&confs=TASLP&confs=TIP&confs=TKDE&confs=TOIS&confs=TPAMI&confs=VLDB&confs=WACV&confs=WSDM&confs=WWW

Wei Wan, Shengshan Hu, Jianrong Lu, LEO YU ZHANG, Hai Jin and Yuanyuan He. Shielding Federated Learning: Robust Aggregation with Adaptive Client Selection

Limitations of Existing Defenses 1.They pay little attention to the client selection step. 2.They are vulnerable to sybil/collusion attacks. 3.They perform poorly in non-IID scenarios. 4.They rely on unrealistic assumptions (e.g.,validation dataset,number of attackers,local dataset size). 289/5000 现有防御的局限性

1.他们很少注意客户选择的步骤。

2.它们很容易受到sybil/勾结攻击。

3.它们在非iid场景中表现不佳。

4.它们依赖于不现实的假设(例如，验证数据集、攻击者的数量、本地数据集的大小)。

Hong Zhang, Ji Liu, Juncheng Jia, Yang Zhou, Huaiyu Dai and Dejing Dou. FedDUAP: Federated Learning with Dynamic Update and Adaptive Pruning Using Shared Data on the Server

数据异构和效率-noiid

–