MDL denoising
- 1 November 2000
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 46 (7) , 2537-2543
- https://doi.org/10.1109/18.887861
Abstract
'The so-called denoising problem, relative to normal models for noise, is formalized such that "noise" is defined as the incompressible part in the data while the compressible part defines the meaningful information-bearing signal. Such a decomposition is effected by minimization of the ideal code length, called for by the minimum description length (MDL) principle, and obtained by an application of the normalized maximum-likelihood technique to the primary parameters, their range, and their number. For any orthonormal regression matrix, such as defined by wavelet transforms, the minimization can be done with a threshold for the squared coefficients resulting from the expansion of the data sequence in the basis vectors defined by the matrix.Keywords
This publication has 9 references indexed in Scilit:
- A generalized minmax bound for universal codingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Elements of Information TheoryPublished by Wiley ,2001
- Model Selection and the Principle of Minimum Description LengthJournal of the American Statistical Association, 2001
- Hypothesis Selection and Testing by the MDL PrincipleThe Computer Journal, 1999
- Minimax description length for signal denoising and optimized representationIEEE Transactions on Information Theory, 1999
- The minimum description length principle in coding and modelingIEEE Transactions on Information Theory, 1998
- Fisher information and stochastic complexityIEEE Transactions on Information Theory, 1996
- The Risk Inflation Criterion for Multiple RegressionThe Annals of Statistics, 1994
- Ideal Spatial Adaptation by Wavelet ShrinkageBiometrika, 1994