AI Paper Atlas

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

Improving Compositional Generalization using Iterated Learning and Simplicial Embeddings

Yi Ren, Samuel Lavoie, Mikhail Galkin, Danica J. Sutherland, Aaron Courville

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 1 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
Add iterated learning to SEM representationsGCN on ogbg-molhiv · ogbg-molhiv Test-fullAUROC: 78.40$\pm$0.67 -> 79.09$\pm$0.67Improved
Add iterated learning to SEM representationsGIN on ogbg-molhiv · ogbg-molhiv Test-fullAUROC: 78.21$\pm$0.78 -> 78.61$\pm$0.73Improved
Add iterated learning to SEM representationsGCN on ogbg-molpcba · ogbg-molpcba Test-fullAverage precision: 25.89$\pm$0.71 -> 27.15$\pm$0.74Improved
Add iterated learning to SEM representationsGIN on ogbg-molpcba · ogbg-molpcba Test-fullAverage precision: 25.66$\pm$0.47 -> 28.02$\pm$0.61Improved
Add iterated learning to SEM representationsGCN on PCQM4Mv2 · PCQM4Mv2 ValidMAE: 0.106$\pm$0.002 -> 0.102$\pm$0.005Improved
Add iterated learning to SEM representationsGIN on PCQM4Mv2 · PCQM4Mv2 ValidMAE: 0.106$\pm$0.004 -> 0.098$\pm$0.005Improved
Use SEM discretization without iterated learning (ablation)3dShapes compositional regression · 3dShapes compositional testtest errorNo clear effect
Use iterated learning with continuous representations instead of SEM (ablation)3dShapes compositional regression · 3dShapes compositional testtest errorNo clear effect
Use argmax pseudo-labels instead of sampling from the teacher distribution (ablation)SEM-IL · 3dShapestest performanceHurt
Add iterated learning to SEM representationsMolecular graph probing on Molhiv-trained representations · Molecular graph probing (Molhiv-trained)Average AUROC: 0.836 -> 0.878Improved
Add iterated learning to SEM representationsMolecular graph probing on Molpcba-trained representations · Molecular graph probing (Molpcba-trained)Average AUROC: 0.926 -> 0.931Improved

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