analiza aiqbench
This paper proposes a way to get more diverse samples from GFlowNets, which are used to generate varied, high-reward candidates when oracle queries are expensive. Instead of scaling a single policy, it trains a population of GFlowNets that divide the state space through a shared classifier and can exchange trajectories. The results are reported on HyperGrid and a genomic design task, with the most meaningful claim being that coverage keeps growing with community size while data-parallel training plateaus. Readers should weigh these as benchmark results from a single 30-page preprint still under review.
Mixture of GFlowNets
arXiv:2610.17007v1 Announce Type: new Abstract: Generative Flow Networks (GFlowNets) learn a policy that samples compositional objects in proportion to an unnormalized reward $R(x)$. This makes them useful wherever many diverse, high-reward candidates are needed and oracle queries are expensive. In practice, a single GFlowNet often fails to cover all modes of a complex distribution, and the obvious ways to scale compute saturate: larger batches and data-parallel gradient averaging draw every trajectory from one, possibly biased, policy, so extra rollouts rediscover known modes instead of finding new ones. We explore whether this mode-bias problem (mode collapse in the extreme) can be addressed with a population of asynchronously trained policies which divide the state space between them. Mixture of GFlowNets (MoG) replaces the single forward policy with a community of $K$ independent GFlowNets coupled through one jointly trained soft classifier, which assigns each member its own region of the state space through reward shaping. Mixing the members under their learned partition functions provably recovers the target $p(x)\propto R(x)$. MoG exchange (MoGX) also lets the classifier route sampled trajectories: a trajectory the classifier assigns to another member is sent to that member and trained on there, with no importance sampling. We evaluate both methods by keeping the environment-interaction budget fixed while scaling the community size, turning parallel compute into mode coverage without extra oracle calls. On HyperGrid, across six reward landscapes and communities of up to $K{=}64$, MoG and MoGX keep finding new modes as $K$ grows, while data-parallel training saturates. On a genomic ideotype-design task, MoGX covers $39%$ of the $1024$ optima at $K{=}64$, against $31\%$ for independent agents and $20\%$ for data-parallel training. MoG also has a lower per-iteration synchronization cost than data-parallel training.
Abstract:Generative Flow Networks (GFlowNets) learn a policy that samples compositional objects in proportion to an unnormalized reward $R(x)$. This makes them useful wherever many diverse, high-reward candidates are needed and oracle queries are expensive. In practice, a single GFlowNet often fails to cover all modes of a complex distribution, and the obvious ways to scale compute saturate: larger batches and data-parallel gradient averaging draw every trajectory from one, possibly biased, policy, so extra rollouts rediscover known modes instead of finding new ones. We explore whether this mode-bias problem (mode collapse in the extreme) can be addressed with a population of asynchronously trained policies which divide the state space between them. Mixture of GFlowNets (MoG) replaces the single forward policy with a community of $K$ independent GFlowNets coupled through one jointly trained soft classifier, which assigns each member its own region of the state space through reward shaping. Mixing the members under their learned partition functions provably recovers the target $p(x)\propto R(x)$. MoG exchange (MoGX) also lets the classifier route sampled trajectories: a trajectory the classifier assigns to another member is sent to that member and trained on there, with no importance sampling. We evaluate both methods by keeping the environment-interaction budget fixed while scaling the community size, turning parallel compute into mode coverage without extra oracle calls. On HyperGrid, across six reward landscapes and communities of up to $K{=}64$, MoG and MoGX keep finding new modes as $K$ grows, while data-parallel training saturates. On a genomic ideotype-design task, MoGX covers $39%$ of the $1024$ optima at $K{=}64$, against $31\%$ for independent agents and $20\%$ for data-parallel training. MoG also has a lower per-iteration synchronization cost than data-parallel training.
30 Pages, 21 Figures, 14 Tables, under review. Omar G. Younis and Joseph D. Viviano are co-lead authors
Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
arXiv:2610.17007 [cs.LG]
(or arXiv:2610.17007v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.17007 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 396 cuvinte
Subiecte conexe
Learning a Mixture of GFlowNets
arXiv · 7 oct. 2026
Beyond Mode Collapse: Generating Diverse Synthetic Expert Conversations via Generative Flow Networks
arXiv · 1 oct. 2026
Introducing Instella-MoE: A State-of-the-Art Fully Open Mixture-of-Experts Language Model
AMD · 24 iul. 2026
Multiclass Classification without Labels via Posterior Simplex Geometry
arXiv · 29 iul. 2026