Pith. sign in

REVIEW 4 cited by

GFlowNets and variational inference

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2210.00580 v3 pith:KULRDHAN submitted 2022-10-02 cs.LG stat.ML

Nikolay Malkin , Salem Lahlou , Tristan Deleu , Xu Ji , Edward Hu , Katie Everett , Dinghuai Zhang , Yoshua Bengio This is my paper · ORCID
classification cs.LGstat.ML
keywords gflownetsdistributionsalgorithmscasesdifferencesfamiliesinferencelearning
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This paper builds bridges between two families of probabilistic algorithms: (hierarchical) variational inference (VI), which is typically used to model distributions over continuous spaces, and generative flow networks (GFlowNets), which have been used for distributions over discrete structures such as graphs. We demonstrate that, in certain cases, VI algorithms are equivalent to special cases of GFlowNets in the sense of equality of expected gradients of their learning objectives. We then point out the differences between the two families and show how these differences emerge experimentally. Notably, GFlowNets, which borrow ideas from reinforcement learning, are more amenable than VI to off-policy training without the cost of high gradient variance induced by importance sampling. We argue that this property of GFlowNets can provide advantages for capturing diversity in multimodal target distributions.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Beyond the Proxy: Trajectory-Distilled Guidance for Offline GFlowNet Training

    cs.LG 2025-05 conditional novelty 6.0 of 10

    TD-GFN uses IRL-derived edge rewards to prune the environment DAG and sample backward trajectories, training offline GFlowNets directly from ground-truth terminal rewards without a proxy reward model.

  2. The Curious Case of the Default Settings: Evaluating Default Performance of Variational Inference Software

    stat.CO 2026-08 conditional novelty 5.0 of 10

    Default settings in PyMC, NumPyro, and TensorFlow Probability can yield biased or silently broken variational inference results even in simple one-dimensional conjugate models.

  3. Diffusion-Augmented Markov Decision Processes for Maximum Entropy Reinforcement Learning

    cs.LG 2025-12 conditional novelty 5.0 of 10

    Diffusion policies can be inserted into maximum-entropy RL by minimizing an upper bound on reverse KL, yielding DiffPPO, DiffSAC, and DiffWPO.

  4. Torsional-GFN: a conditional conformation generator for small molecules

    cs.LG 2025-07 conditional novelty 5.0 of 10

    Torsional-GFN, a conditional GFlowNet with a new graph network, samples torsion angles of small molecules to approximate the Boltzmann distribution, with partial generalization to unseen local structures.

Pith tools