AI Paper Atlas

Research map / Federated learning and privacy / Privacy attacks and secure learning

Chameleon: Adapting to Peer Images for Planting Durable Backdoors in Federated Learning

Yanbo Dai, Songze Li

ICML 2023 · Federated learning and privacy › Privacy attacks and secure learning · approach: Defending federated learning against poisoning · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use Chameleon contrastive adaptation and classifier projection instead of prior backdoor attacksFederated visual backdoor training · CIFAR100 bicycle with black background50%-Lifespan: 229 -> 661Improved
Use Chameleon instead of prior attacksFederated visual backdoor training · CIFAR10 car with striped background50%-Lifespan: 283 -> 1106Improved
Use Chameleon instead of prior attacksFederated visual backdoor training · CIFAR10 southwest airline edge-case60%-Lifespan: 339 vs 349Mixed
Use Chameleon rather than BaselineCIFAR100 bicycle with black background under weakly DP and NCD · CIFAR10050%-Lifespan: 150 vs -Improved
Use Chameleon rather than NeurotoxinCIFAR100 bicycle with black background under Krum and NCD · CIFAR10050%-Lifespan: 53 -> 72Improved
Add Chameleon’s peer-image-aware trainingCIFAR100 bicycle with black background in FedProx · FedProx, Dirichlet α=0.950%-Lifespan: 472 -> 1522Improved
Increase pattern diffusion from p_d=0.1 to p_d=0.5 (ablation)Chameleon · CIFAR10 Type-2 pixel-pattern backdoor50%-Lifespan: 833(50%) -> 720(50%)Hurt
Increase β from 3 to 12 (ablation)Chameleon pixel-pattern backdoor · Type-2 pixel-pattern, p_d=0.5Backdoor durabilityImproved
Vary β, with the strongest result at an intermediate β=4 (ablation)Chameleon semantic backdoor · CIFAR10 car-with-racing-stripesBackdoor durability: β=4Mixed
Remove interferers from subsequent benign-client datasets (ablation)Backdoor trained with model replacement · CIFAR10 car-to-bird semantic backdoorBackdoor accuracy persistenceImproved
Remove facilitators from subsequent benign-client datasets (ablation)Backdoor trained with model replacement · CIFAR10 car-to-bird semantic backdoorBackdoor durabilityHurt
Train using only poisoned images rather than mixing in benign images (ablation)Malicious local backdoor model · CIFAR10 car-to-bird semantic backdoorBackdoor lifespanHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy