Limitations on Robust Ratings and Predictions

Abstract : Predictions are a well-studied form of ratings. Their objective nature allows a rigourous analysis. A problem is that there are attacks on prediction systems and rating systems. These attacks decrease the usefulness of the predictions. Attackers may ignore the incentives in the system, so we may not rely on these to protect ourselves. The user must block attackers, ideally before the attackers introduce too much misinformation. We formally axiomatically define robustness as the property that no rater can introduce too much misinformation. We formally prove that notions of robustness come at the expense of other desirable properties, such as the lack of bias or effectiveness. We also show that there do exist trade-offs between the different properties, allowing a prediction system with limited robustness, limited bias and limited effectiveness.
Document type :
Conference papers
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.inria.fr/hal-01438352
Contributor : Hal Ifip <>
Submitted on : Tuesday, January 17, 2017 - 4:08:10 PM
Last modification on : Tuesday, January 17, 2017 - 4:18:27 PM
Long-term archiving on : Tuesday, April 18, 2017 - 3:18:47 PM

File

428098_1_En_8_Chapter.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Tim Muller, Yang Liu, Jie Zhang. Limitations on Robust Ratings and Predictions. 10th IFIP International Conference on Trust Management (TM), Jul 2016, Darmstadt, Germany. pp.113-126, ⟨10.1007/978-3-319-41354-9_8⟩. ⟨hal-01438352⟩

Share

Metrics

Record views

85

Files downloads

144