arXiv Machine Learning

Particle GFlowNets: Rethinking Generative Marginalization Models

The paper introduces Particle GFlowNets, showing that Generative Marginalization Models (MaMs) are equivalent to Generative Flow Networks. It extends MaMs to non‑autoregressive sampling and proposes an automatic full‑state rejuvenation criterion based on the Gelman‑Rubin statistic to accelerate learning. Experiments demonstrate significant training speedups in large combinatorial spaces.

arXiv Machine Learning
Aug 31

Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Network

The paper introduces JSP-GFN, a Generative Flow Network that jointly infers the structure and parameters of a Bayesian Network. It sequentially generates a directed acyclic graph edge by edge and then samples the corresponding conditional probability parameters once the full structure is known. Experiments on simulated and real data show that JSP‑GFN accurately approximates the joint posterior and outperforms existing methods.

By Tristan Deleu, Mizu Nishikawa-Toomey, Jithendaraa Subramanian, Esmeralda S. Whitammer, Laurent Charlin, Yoshua Bengio
arXiv Machine Learning
Oct 16

Conditional Mean Flows: Few-Step Densities for Simulation-Based Inference

Conditional Mean Flows introduce a new approach for simulation-based inference that replaces the costly numerical integration of traditional flow matching with few-step sampling and tractable density evaluation. By identifying conditional flow maps through a mean flow objective and using independent masking, the method allows direct estimation of marginal likelihoods. Experiments demonstrate that single-step conditional mean flow density evaluation achieves error comparable to 200-step flow matching while accelerating downstream MCMC sampling by roughly 200‑fold.

By Jonas Arruda, Jan Hasenauer, Stefan T. Radev
arXiv Machine Learning
Jul 7

Non-Asymptotic Error Bounds for SMC with Biased Proposals: Application to Conditional Diffusion Sampling

arXiv:2607. 04780v1 Announce Type: cross Abstract: Sequential Monte Carlo (SMC) methods are a natural tool for post-hoc conditioning of pretrained generative models, but in many applications the mutation kernels used by the particle system are biased approximations of an ideal Feynman--Kac flow.

By Stanislas Strasman (SU, LPSM), Gabriel Victorino Cardoso (LPSM), Sylvain Le Corff (LPSM), Vincent Lemaire (LPSM), Antonio Ocello
arXiv Machine Learning
Sep 1

Delta-AI: Local objectives for amortized inference in sparse graphical models

arXiv:2310.02423v3 Announce Type: replace Abstract: We present a new algorithm for amortized inference in sparse probabilistic graphical models (PGMs), which we call $\Delta$-amortized inference ($\D...

By Jean-Pierre Falet, Hae Beom Lee, Esmeralda S. Whitammer, Chen Sun, Dragos Secrieru, Thomas Jiralerspong, Dinghuai Zhang, Guillaume Lajoie, Yoshua Bengio
arXiv Statistics ML
Nov 9

Training-Free Generative Sampling via Moment-Matched Score Smoothing

The paper introduces MM‑SOLD, a training‑free sampling method that uses moment‑matched score‑smoothed overdamped Langevin dynamics to generate samples without training a neural diffusion model. It enforces target moments during sampling, proving convergence to a Gibbs–Boltzmann density that matches the mean and covariance of a training Gaussian mixture model, even for rank‑deficient data. Experiments demonstrate fast, robust latent‑space image generation on CPUs with fidelity and diversity comparable to neural diffusion baselines.

By Zhenyu Yao, Daniel Paulin