AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Agnostic Learning with Multiple Objectives

Corinna Cortes, Mehryar Mohri, Javier Gonzalvo, Dmitry Storcheus

NeurIPS 2020 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 8 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
Replace uniform mixture weights with ALMOLogistic regression · MNISTcross-entropy loss: 0.1412 -> 0.0726Improved
Replace uniform weights with ALMOLogistic regression · Adulthinge loss: 0.967 -> 0.518Improved
Replace uniform weights with ALMOLogistic regression · Adultcross-entropy loss: 0.680 -> 0.501Improved
Replace uniform weights with ALMOLogistic regression · Fashion MNISThinge loss: 0.633 -> 0.695Hurt
Replace uniform weights with ALMO (1-norm loss gets low weight)Logistic regression · MNIST1-norm loss: 0.2349 -> 0.6587Hurt
Replace uniform weights with ALMOLogistic regression · AdultAUC: 0.8291 -> 0.8321Improved
Replace uniform weights with ALMOLogistic regression · MNISTAUC: 0.9801 -> 0.9877Improved
Replace uniform weights with ALMODNN · Adulthinge loss: 0.914 -> 0.389Improved
Replace uniform weights with ALMODNN · AdultAUC: 0.8092 -> 0.8840Improved
Replace uniform weights with ALMODNN · MNISTcross-entropy loss: 0.044 -> 0.045No clear effect
Replace uniform weights with ALMODNN · MNISThinge loss: 0.631 -> 0.752Hurt
Replace uniform weights with ALMODNN · Fashion MNISThinge loss: 0.661 -> 0.547Improved

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