Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310263258> ?p ?o ?g. }
Showing items 1 to 51 of
51
with 100 items per page.
- W4310263258 abstract "We consider the general problem of recovering a high-dimensional signal from noisy quantized measurements. Quantization, especially coarse quantization such as 1-bit sign measurements, leads to severe information loss and thus a good prior knowledge of the unknown signal is helpful for accurate recovery. Motivated by the power of score-based generative models (SGM, also known as diffusion models) in capturing the rich structure of natural signals beyond simple sparsity, we propose an unsupervised data-driven approach called quantized compressed sensing with SGM (QCS-SGM), where the prior distribution is modeled by a pre-trained SGM. To perform posterior sampling, an annealed pseudo-likelihood score called noise perturbed pseudo-likelihood score is introduced and combined with the prior score of SGM. The proposed QCS-SGM applies to an arbitrary number of quantization bits. Experiments on a variety of baseline datasets demonstrate that the proposed QCS-SGM significantly outperforms existing state-of-the-art algorithms by a large margin for both in-distribution and out-of-distribution samples. Moreover, as a posterior sampling method, QCS-SGM can be easily used to obtain confidence intervals or uncertainty estimates of the reconstructed results. The code is available at https://github.com/mengxiangming/QCS-SGM." @default.
- W4310263258 created "2022-11-30" @default.
- W4310263258 creator A5002600110 @default.
- W4310263258 creator A5078150685 @default.
- W4310263258 date "2022-11-02" @default.
- W4310263258 modified "2023-09-25" @default.
- W4310263258 title "Quantized Compressed Sensing with Score-based Generative Models" @default.
- W4310263258 doi "https://doi.org/10.48550/arxiv.2211.13006" @default.
- W4310263258 hasPublicationYear "2022" @default.
- W4310263258 type Work @default.
- W4310263258 citedByCount "0" @default.
- W4310263258 crossrefType "posted-content" @default.
- W4310263258 hasAuthorship W4310263258A5002600110 @default.
- W4310263258 hasAuthorship W4310263258A5078150685 @default.
- W4310263258 hasBestOaLocation W43102632581 @default.
- W4310263258 hasConcept C107673813 @default.
- W4310263258 hasConcept C11413529 @default.
- W4310263258 hasConcept C119857082 @default.
- W4310263258 hasConcept C153180895 @default.
- W4310263258 hasConcept C154945302 @default.
- W4310263258 hasConcept C177769412 @default.
- W4310263258 hasConcept C28855332 @default.
- W4310263258 hasConcept C33923547 @default.
- W4310263258 hasConcept C41008148 @default.
- W4310263258 hasConcept C774472 @default.
- W4310263258 hasConceptScore W4310263258C107673813 @default.
- W4310263258 hasConceptScore W4310263258C11413529 @default.
- W4310263258 hasConceptScore W4310263258C119857082 @default.
- W4310263258 hasConceptScore W4310263258C153180895 @default.
- W4310263258 hasConceptScore W4310263258C154945302 @default.
- W4310263258 hasConceptScore W4310263258C177769412 @default.
- W4310263258 hasConceptScore W4310263258C28855332 @default.
- W4310263258 hasConceptScore W4310263258C33923547 @default.
- W4310263258 hasConceptScore W4310263258C41008148 @default.
- W4310263258 hasConceptScore W4310263258C774472 @default.
- W4310263258 hasLocation W43102632581 @default.
- W4310263258 hasOpenAccess W4310263258 @default.
- W4310263258 hasPrimaryLocation W43102632581 @default.
- W4310263258 hasRelatedWork W1978450727 @default.
- W4310263258 hasRelatedWork W2033914206 @default.
- W4310263258 hasRelatedWork W2146076056 @default.
- W4310263258 hasRelatedWork W2163831990 @default.
- W4310263258 hasRelatedWork W2344532017 @default.
- W4310263258 hasRelatedWork W2378160586 @default.
- W4310263258 hasRelatedWork W2965938919 @default.
- W4310263258 hasRelatedWork W3003836766 @default.
- W4310263258 hasRelatedWork W3107474891 @default.
- W4310263258 hasRelatedWork W4244943737 @default.
- W4310263258 isParatext "false" @default.
- W4310263258 isRetracted "false" @default.
- W4310263258 workType "article" @default.