AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

Comparing Apples to Oranges: Learning Similarity Functions for Data Produced by Different Distributions

Leonidas Tsepenekas, Ivan Brugere, Freddy Lecue, Daniele Magazzeni

NeurIPS 2023 · Learning theory and algorithms › Clustering · approach: Clustering approximation algorithms · 0 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 Naive algorithm instead of XGB regressorsimilarity learning across groups · Credit Card Defaultaverage relative error (%): 3.121 -> 1.558Improved
Use Naive algorithm instead of best baseline XGBsimilarity learning across groups · Adultaverage relative error (%): 2.248 -> 1.319Improved
Use Naive algorithm instead of best baseline XGBsimilarity learning across groups · Give Me Some Creditaverage relative error (%): 5.614 -> 1.630Improved
Use Naive algorithm instead of MLPsimilarity learning across groups · Credit Card Defaultaverage absolute error / epsilon: 1.465 -> 0.636Improved
Use Naive algorithm instead of RFsimilarity learning across groups · Give Me Some Creditaverage absolute error / epsilon: 7.258 -> 2.644Improved
Use Naive algorithm instead of RFsimilarity learning across groups · Give Me Some CreditSD of relative error (%): 12.404 -> 4.106Improved
Use Cluster instead of Naive (fewer queries)two-phase similarity learner · Credit Card Defaultoracle queries: 81.01Improved
Use Cluster instead of Naive (fewer queries)two-phase similarity learner · Adultoracle queries: 80.40Improved
Use Cluster instead of Naive (fewer queries)two-phase similarity learner · Give Me Some Creditoracle queries: 84.87Improved
Use Cluster instead of Naivetwo-phase similarity learner · Credit Card Defaultaverage relative error (%): 1.558 -> 1.593No clear effect
Use Cluster instead of Naivetwo-phase similarity learner · Adultaverage relative error (%): 1.319 -> 1.321No clear effect
Set dilation factor to 2 instead of 8 in Cluster training (ablation)Cluster · synthetic similarityaccuracy vs queries balanceImproved

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