High-accuracy sampling for diffusion models and log-concave distributions
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917438165090304 |
|---|---|
| author | Chen, Fan Chewi, Sinho Daskalakis, Constantinos Rakhlin, Alexander |
| author_facet | Chen, Fan Chewi, Sinho Daskalakis, Constantinos Rakhlin, Alexander |
| contents | We present algorithms for diffusion model sampling which obtain $δ$-error in $\mathrm{polylog}(1/δ)$ steps, given access to $\widetilde O(δ)$-accurate score estimates in $L^2$. This is an exponential improvement over all previous results. Specifically, under minimal data assumptions, the complexity is $\widetilde O(d_\star \mathrm{polylog}(1/δ))$ where $d_\star$ is the intrinsic dimension of the data. Further, under a non-uniform $L$-Lipschitz condition, the complexity reduces to $\widetilde O(L \mathrm{polylog}(1/δ))$. Our approach also yields the first $\mathrm{polylog}(1/δ)$ complexity sampler for general log-concave distributions using only gradient evaluations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_01338 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | High-accuracy sampling for diffusion models and log-concave distributions Chen, Fan Chewi, Sinho Daskalakis, Constantinos Rakhlin, Alexander Machine Learning Statistics Theory We present algorithms for diffusion model sampling which obtain $δ$-error in $\mathrm{polylog}(1/δ)$ steps, given access to $\widetilde O(δ)$-accurate score estimates in $L^2$. This is an exponential improvement over all previous results. Specifically, under minimal data assumptions, the complexity is $\widetilde O(d_\star \mathrm{polylog}(1/δ))$ where $d_\star$ is the intrinsic dimension of the data. Further, under a non-uniform $L$-Lipschitz condition, the complexity reduces to $\widetilde O(L \mathrm{polylog}(1/δ))$. Our approach also yields the first $\mathrm{polylog}(1/δ)$ complexity sampler for general log-concave distributions using only gradient evaluations. |
| title | High-accuracy sampling for diffusion models and log-concave distributions |
| topic | Machine Learning Statistics Theory |
| url | https://arxiv.org/abs/2602.01338 |