AI & ML
Elliptic Sampling: Optimal Interpolant Schedules for Gaussians and Beyond
arXiv stat.ML0 views
arXiv:2611.18542v1 Announce Type: new Abstract: Generative modeling based on stochastic interpolants requires specifying two key choices: an interpolant schedule bridging the initial and target distributions, and an efficient discretization of the resulting probability-flow ODE. In this work, we introduce a new framework for joint schedule/discretization design based on the analyticity properties of the (complexified) ODE trajectories, which in turn ensure good numerical behavior. Our formulation can be solved exactly via conformal mapping, yielding a natural family of variance-preserving elliptic schedules, parametrized by Jacobi elliptic functions. Pairing an elliptic schedule with a tailored discretization, we obtain Zoloflow, a new sampling algorithm closely connected to Zolotarev's optimal rational function approximants. Crucially, both the elliptic schedule and Zoloflow explicitly incorporate the condition number $\kappa$ of the target distribution and reduce to the classical trig schedule in the limit $\kappa \rightarrow 1$. For Gaussian targets, Zoloflow achieves minimax optimality among all oblivious score-span algorithms, attaining total-variation error $\varepsilon$ using $\Theta(\log\kappa\,\log(\sqrt d/\varepsilon))$ (exact) score queries. For the log-concave case, the elliptic schedule allows sampling with at most $\varepsilon$ 2-Wasserstein-error by requiring merely $\widetilde{O}(\sqrt{d} \, \mathrm{polylog}(\kappa, 1/\varepsilon))$ score evaluations. Complementing these theoretical results, we empirically evaluate the elliptic schedule on general (non-log-concave) target distributions, including detailed comparisons with alternative schedule proposals on both synthetic benchmarks and existing image generation datasets.
Source:https://arxiv.org/abs/2611.18542PreviousBregman geometry in regular exponential families: dual representations and relative smoothnessNextDistributionally Robust Risk Evaluation with Low-Rank Perturbations
Related posts
Digital Fairness Act
0Hacker News New
Show HN: Furnish – AI-powered home design with editable 2D plans and 3D views
0Hacker News New
Ask HN: How are you dealing with AI burnout?
0Hacker News Ask
May be data is the bottleneck for AGI
0Hacker News Ask
Show HN: Otus – EU web analytics with a Mistral powered AI analyst
0Hacker News Show
Show HN: Deslop skill based on corpus-linguistics and psycholinguistics research
0Hacker News Show