A two-term penalty function for inverse problems with sparsity constrains. Rodriguez, P. In 2017 25th European Signal Processing Conference (EUSIPCO), pages 2126-2130, Aug, 2017.
![A two-term penalty function for inverse problems with sparsity constrains [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex
Paper doi abstract bibtex Inverse problems with sparsity constrains, such Basis Pursuit denoising (BPDN) and Convolutional BPDN (CBPDN), usually use the ℓ1-norm as the penalty function; however such choice leads to a solution that is biased towards zero. Recently, several works have proposed and assessed the properties of other non-standard penalty functions (most of them non-convex), which avoid the above mentioned drawback and at the same time are intended to induce sparsity more strongly than the ℓ1-norm. In this paper we propose a two-term penalty function consisting of a synthesis between the ℓ1-norm and the penalty function associated with the Non-Negative Garrote (NNG) thresholding rule. Although the proposed two-term penalty function is non-convex, the total cost function for the BPDN / CBPDN problems is still convex. The performance of the proposed two-term penalty function is compared with other reported choices for practical denoising, deconvolution and convolutional sparse coding (CSC) problems within the BPDN / CBPDN frameworks. Our experimental results show that the proposed two-term penalty function is particularly effective (better reconstruction with sparser solutions) for the CSC problem while attaining competitive performance for the denoising and deconvolution problems.
@InProceedings{8081585,
author = {P. Rodriguez},
booktitle = {2017 25th European Signal Processing Conference (EUSIPCO)},
title = {A two-term penalty function for inverse problems with sparsity constrains},
year = {2017},
pages = {2126-2130},
abstract = {Inverse problems with sparsity constrains, such Basis Pursuit denoising (BPDN) and Convolutional BPDN (CBPDN), usually use the ℓ1-norm as the penalty function; however such choice leads to a solution that is biased towards zero. Recently, several works have proposed and assessed the properties of other non-standard penalty functions (most of them non-convex), which avoid the above mentioned drawback and at the same time are intended to induce sparsity more strongly than the ℓ1-norm. In this paper we propose a two-term penalty function consisting of a synthesis between the ℓ1-norm and the penalty function associated with the Non-Negative Garrote (NNG) thresholding rule. Although the proposed two-term penalty function is non-convex, the total cost function for the BPDN / CBPDN problems is still convex. The performance of the proposed two-term penalty function is compared with other reported choices for practical denoising, deconvolution and convolutional sparse coding (CSC) problems within the BPDN / CBPDN frameworks. Our experimental results show that the proposed two-term penalty function is particularly effective (better reconstruction with sparser solutions) for the CSC problem while attaining competitive performance for the denoising and deconvolution problems.},
keywords = {approximation theory;convex programming;image coding;image denoising;inverse problems;optimisation;signal denoising;signal reconstruction;two-term penalty function;inverse problems;sparsity constrains;nonstandard penalty functions;BPDN-CBPDN problems;convolutional sparse coding problems;convolutional basis pursuit denoising;Signal processing algorithms;Convergence;Noise reduction;Convolutional codes;Convolution;Dictionaries;Europe},
doi = {10.23919/EUSIPCO.2017.8081585},
issn = {2076-1465},
month = {Aug},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2017/papers/1570346716.pdf},
}Downloads: 0
{"_id":"ovjPTXk7nn9tiJePX","bibbaseid":"rodriguez-atwotermpenaltyfunctionforinverseproblemswithsparsityconstrains-2017","authorIDs":[],"author_short":["Rodriguez, P."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["P."],"propositions":[],"lastnames":["Rodriguez"],"suffixes":[]}],"booktitle":"2017 25th European Signal Processing Conference (EUSIPCO)","title":"A two-term penalty function for inverse problems with sparsity constrains","year":"2017","pages":"2126-2130","abstract":"Inverse problems with sparsity constrains, such Basis Pursuit denoising (BPDN) and Convolutional BPDN (CBPDN), usually use the ℓ1-norm as the penalty function; however such choice leads to a solution that is biased towards zero. Recently, several works have proposed and assessed the properties of other non-standard penalty functions (most of them non-convex), which avoid the above mentioned drawback and at the same time are intended to induce sparsity more strongly than the ℓ1-norm. In this paper we propose a two-term penalty function consisting of a synthesis between the ℓ1-norm and the penalty function associated with the Non-Negative Garrote (NNG) thresholding rule. Although the proposed two-term penalty function is non-convex, the total cost function for the BPDN / CBPDN problems is still convex. The performance of the proposed two-term penalty function is compared with other reported choices for practical denoising, deconvolution and convolutional sparse coding (CSC) problems within the BPDN / CBPDN frameworks. Our experimental results show that the proposed two-term penalty function is particularly effective (better reconstruction with sparser solutions) for the CSC problem while attaining competitive performance for the denoising and deconvolution problems.","keywords":"approximation theory;convex programming;image coding;image denoising;inverse problems;optimisation;signal denoising;signal reconstruction;two-term penalty function;inverse problems;sparsity constrains;nonstandard penalty functions;BPDN-CBPDN problems;convolutional sparse coding problems;convolutional basis pursuit denoising;Signal processing algorithms;Convergence;Noise reduction;Convolutional codes;Convolution;Dictionaries;Europe","doi":"10.23919/EUSIPCO.2017.8081585","issn":"2076-1465","month":"Aug","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2017/papers/1570346716.pdf","bibtex":"@InProceedings{8081585,\n author = {P. Rodriguez},\n booktitle = {2017 25th European Signal Processing Conference (EUSIPCO)},\n title = {A two-term penalty function for inverse problems with sparsity constrains},\n year = {2017},\n pages = {2126-2130},\n abstract = {Inverse problems with sparsity constrains, such Basis Pursuit denoising (BPDN) and Convolutional BPDN (CBPDN), usually use the ℓ1-norm as the penalty function; however such choice leads to a solution that is biased towards zero. Recently, several works have proposed and assessed the properties of other non-standard penalty functions (most of them non-convex), which avoid the above mentioned drawback and at the same time are intended to induce sparsity more strongly than the ℓ1-norm. In this paper we propose a two-term penalty function consisting of a synthesis between the ℓ1-norm and the penalty function associated with the Non-Negative Garrote (NNG) thresholding rule. Although the proposed two-term penalty function is non-convex, the total cost function for the BPDN / CBPDN problems is still convex. The performance of the proposed two-term penalty function is compared with other reported choices for practical denoising, deconvolution and convolutional sparse coding (CSC) problems within the BPDN / CBPDN frameworks. Our experimental results show that the proposed two-term penalty function is particularly effective (better reconstruction with sparser solutions) for the CSC problem while attaining competitive performance for the denoising and deconvolution problems.},\n keywords = {approximation theory;convex programming;image coding;image denoising;inverse problems;optimisation;signal denoising;signal reconstruction;two-term penalty function;inverse problems;sparsity constrains;nonstandard penalty functions;BPDN-CBPDN problems;convolutional sparse coding problems;convolutional basis pursuit denoising;Signal processing algorithms;Convergence;Noise reduction;Convolutional codes;Convolution;Dictionaries;Europe},\n doi = {10.23919/EUSIPCO.2017.8081585},\n issn = {2076-1465},\n month = {Aug},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2017/papers/1570346716.pdf},\n}\n\n","author_short":["Rodriguez, P."],"key":"8081585","id":"8081585","bibbaseid":"rodriguez-atwotermpenaltyfunctionforinverseproblemswithsparsityconstrains-2017","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2017/papers/1570346716.pdf"},"keyword":["approximation theory;convex programming;image coding;image denoising;inverse problems;optimisation;signal denoising;signal reconstruction;two-term penalty function;inverse problems;sparsity constrains;nonstandard penalty functions;BPDN-CBPDN problems;convolutional sparse coding problems;convolutional basis pursuit denoising;Signal processing algorithms;Convergence;Noise reduction;Convolutional codes;Convolution;Dictionaries;Europe"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2017url.bib","creationDate":"2021-02-13T16:38:25.743Z","downloads":0,"keywords":["approximation theory;convex programming;image coding;image denoising;inverse problems;optimisation;signal denoising;signal reconstruction;two-term penalty function;inverse problems;sparsity constrains;nonstandard penalty functions;bpdn-cbpdn problems;convolutional sparse coding problems;convolutional basis pursuit denoising;signal processing algorithms;convergence;noise reduction;convolutional codes;convolution;dictionaries;europe"],"search_terms":["two","term","penalty","function","inverse","problems","sparsity","constrains","rodriguez"],"title":"A two-term penalty function for inverse problems with sparsity constrains","year":2017,"dataSources":["2MNbFYjMYTD6z7ExY","uP2aT6Qs8sfZJ6s8b"]}