Global-Order GFlowNets

Lluís Pastor-Pérez, Javier Alonso Garcia, Lukas Mauch

Keywords: GFlowNets, Multi-Objective Optimization, XAI, Probabilistic AI, Generative AI, Reinforcement Learning

TL;DR: The paper introduces Global-Order GFlowNets, a new approach for multi-objective optimization that addresses the limitations of Order-Preserving GFlowNets (OP-GFNs), which suffer from conflicting objectives due to local ordering, with a global order.

Abstract:

Order-Preserving (OP) GFlowNets have demonstrated remarkable success in tackling complex multi-objective (MOO) black-box optimization problems using stochastic optimization techniques. Specifically, they can be trained online to efficiently sample diverse candidates near the Pareto front. A key advantage of OP GFlowNets is their ability to impose a local order on training samples based on
Pareto dominance, eliminating the need for scalarization – a common requirement in other approaches like Preference-Conditional GFlowNets. However, we identify an important limitation of OP GFlowNets: imposing a local order on training samples can lead to conflicting optimization objectives. To address this issue, we introduce Global-Order GFlowNets, which transform the local order into a global one, thereby resolving these conflicts. Our experimental evaluations on various benchmarks demonstrate the efficacy and promise of our proposed method.

Submission Number: 31