Strategic clients can inflate their Shapley and leave-one-out data values in federated learning by altering local data during retraining; the proposed Truth-Shapley metric blocks this by ignoring all partial client subsets.
Detection of influential observation in li near regression
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CR 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Data Overvaluation Attack and Truthful Data Valuation in Federated Learning
Strategic clients can inflate their Shapley and leave-one-out data values in federated learning by altering local data during retraining; the proposed Truth-Shapley metric blocks this by ignoring all partial client subsets.