AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Evaluating Recommender System Stability with Influence-Guided Fuzzing

David Shriver, Sebastian Elbaum, Matthew B. Dwyer, David S. Rosenblum

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Recommendation representation learning · 10 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
guide fuzzing with influence models via AMR (rating influence) instead of random additionsUser-User recommender · MovieLens 100k, 1 modificationTopOut instability ratio: 128xImproved
use AMR instead of random additionsFunkSVD · MovieLens 100k, 10 modificationsTopOut instability ratio: over 40000xImproved
use AMA (attribute influence) instead of random additionsLightFM · MovieLens 100k, 10 and 100 modificationsTopOut instability ratio: 1.3x and 3.2xImproved
use AMA instead of random additionsUser-User, Item-Item, FunkSVD (no attribute use) · MovieLens 100kTopOut instabilityNo clear effect
use AMU, ALI, AMR instead of random additionsLightFM · MovieLens 100kTopOut instabilityNo clear effect
use remove-type RMU and change-type CMU instead of randomItem-Item · MovieLens 100kTopOut instability ratio: 6.3x and 13.4xImproved
use Add-type influence heuristics instead of random additionsItem-Item · MovieLens 100kTopOut instabilityNo clear effect
use APORTFOLIO round-robin of four heuristics instead of ARANDUser-User · MovieLens 100k, 100 modificationsmean TopOut instability: 0.031294 -> 0.257794Improved
use APORTFOLIO instead of ARANDFunkSVD · MovieLens 100k, 100 modificationsmean TopOut instability: 0.053446 -> 0.768982Improved
use APORTFOLIO instead of ARANDItem-Item · MovieLens 100k, 100 modificationsmean TopOut instability: 0.026681 -> 0.008653Hurt
use APORTFOLIO instead of ARANDLightFM · MovieLens 100k, 100 modificationsmean TopOut instability: 0.010933 -> 0.017678Improved

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