Unbiased Offline Evaluation of Contextual-bandit-based News Article Recommendation Algorithms

Li, L., Chu, W., Langford, J., & Wang, X. (2011). Unbiased Offline Evaluation of Contextual-bandit-based News Article Recommendation Algorithms. ACM, 297-306.

Type

Conference paper

Intellectual base

Primary research

Year

2011

Pages

297-306

DOI

10.1145/1935826.1935878

DOI registrant

Association for Computing Machinery

Link

https://doi.org/10.1145/1935826.1935878

Cited by

1 citation across 1 artifact.

Each citation links to the sentence it supports in the citing article.

Mechanisms

• Off-Policy or Historical Replay Evaluation
  • Its failure mode is the positivity (overlap) assumption

    This sourceAssumes randomized logging over the available actions and evaluates a policy only on events where its chosen action matches the logged action.

Verification

Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.

Does it back the claim? Not recorded. The single citation of this work carries no recorded support check.

Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.

See how references were verified.

Registry ID ref:3ccf1090d7d8 · see in the full table