AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Multi-Objective GFlowNets

Moksh Jain, Sharath Chandra Raparthy, Alex Hernández-García, Jarrid Rector-Brooks, Yoshua Bengio, Santiago Miret, Emmanuel Bengio

ICML 2023 · Learning theory and algorithms › Combinatorial and search optimization · approach: Multi-objective optimization · 3 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 preference-conditional GFlowNet sampling instead of MOReinforce3 Bigrams · 3 BigramsHV: 0.015 -> 0.220Improved
Use preference-conditional GFlowNet sampling instead of MOReinforceQM9 · QM9Diversity: 0.53 ± 0.08 -> 0.93 ± 0.00Improved
Use MOGFN-PC instead of MOReinforceDNA sequence generation · DNA sequence generationHV: 0.629 ± 0.002 -> 0.517 ± 0.006Hurt
Use MOGFN-PC instead of MOReinforceDNA sequence generation · DNA sequence generationDiversity: 0.6178 ± 0.209 -> 18.131 ± 0.981Improved
Use uniform preference sampling, Dirichlet(α=1), instead of skewed Dirichlet preferences (ablation)MOGFN-PC · 3 Bigrams and fragment-based molecule generationNormalized performanceImproved
Use weighted-sum scalarization instead of weighted-log-sum or weighted Tchebycheff (ablation)MOGFN-PC · 3 Bigrams and fragment-based molecule generationNormalized performanceImproved
Increase reward exponent β (ablation)MOGFN-PC · Tasks in the β analysisPareto performanceMixed
Increase reward exponent β (ablation)MOGFN-PC · Tasks in the β analysisDiversityHurt
Increase policy model size (ablation)MOGFN-PC · 3 BigramsHV: 0.22 -> 0.255 -> 0.262 -> 0.270Improved
Increase policy model size (ablation)MOGFN-PC · 3 BigramsDiversity: 19.79 -> 17.13 -> 17.53 -> 16.12Hurt
Use larger models rather than smaller models (ablation)MOGFN-PC · 3 BigramsReward: 0.44 -> 0.47 -> 0.49 -> 0.51Improved
Use preference thermometer encoding instead of raw preference values (ablation)MOGFN-PC · QM9PerformanceNo clear effect

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