Pith. sign in

REVIEW 6 cited by

An optimal control perspective on diffusion-based generative modeling

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 2211.01364 v3 pith:UGZBVFKS submitted 2022-11-02 cs.LG math.OCstat.ML

Julius Berner , Lorenz Richter , Karen Ullrich This is my paper
classification cs.LGmath.OCstat.ML
keywords controldiffusion-basedgenerativemodelingoptimaldiffusionmodelsperspective
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We establish a connection between stochastic optimal control and generative models based on stochastic differential equations (SDEs), such as recently developed diffusion probabilistic models. In particular, we derive a Hamilton-Jacobi-Bellman equation that governs the evolution of the log-densities of the underlying SDE marginals. This perspective allows to transfer methods from optimal control theory to generative modeling. First, we show that the evidence lower bound is a direct consequence of the well-known verification theorem from control theory. Further, we can formulate diffusion-based generative modeling as a minimization of the Kullback-Leibler divergence between suitable measures in path space. Finally, we develop a novel diffusion-based method for sampling from unnormalized densities -- a problem frequently occurring in statistics and computational sciences. We demonstrate that our time-reversed diffusion sampler (DIS) can outperform other diffusion-based sampling approaches on multiple numerical examples.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

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

  1. From Open Loop to Closed Loop: A Test-Time Iterative Optimization Framework for Reference-Consistent Image Generation

    cs.CV 2026-07 conditional novelty 6.5 of 10

    A training-free closed-loop PID controller iteratively corrects latent control signals so diffusion models stay consistent with ID, pose, or depth references better than matched open-loop sampling.

  2. Stochastic Quantization as Optimal Control

    hep-lat 2026-07 conditional novelty 6.0 of 10

    Stochastic quantization is re-expressed as finite-time optimal control, in which a learned Doob force plus exact path weights reach the Gibbs measure without waiting for equilibrium.

  3. FES-FM: Free Energy Surface Sampling via Reduced Flow Matching

    cs.LG 2026-05 unverdicted novelty 6.0 of 10

    FES-FM learns a reduced flow-matching transport in collective-variable space to sample free energy surfaces, cutting per-sample generation cost while leaving full-space training cost unchanged.

  4. Solving Inverse Problems with Flow-based Models via Model Predictive Control

    eess.IV 2026-01 conditional novelty 6.0 of 10

    MPC-Flow applies model predictive control to guide pretrained flow models through inverse problems, with a single-step variant that avoids backpropagation and scales to 32B-parameter models on consumer hardware.

  5. 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.

  6. Importance Weighted Score Matching for Diffusion Samplers with Enhanced Mode Coverage

    cs.LG 2025-05 conditional novelty 5.0 of 10

    Importance Weighted Score Matching trains diffusion samplers by reweighting score matching with self-normalized importance sampling to approximate the forward KL and improve mode coverage.

Pith tools