Correntropy: Properties and Applications in Non-Gaussian Signal Processing

By Weifeng Liu, Puskal P. Pokharel, José Carlos Príncipe. IEEE Transactions on Signal Processing, 2007

1,555 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/c22ca1c712bc18915dc9737e9d61b69574455021

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. Dacheng Tao
2. Tongliang Liu
3. Ran He
4. Tien-Ping Tan
5. Weishi Zheng
6. Bao-Gang Hu
7. Haiquan Zhao
8. Badong Chen

In Signal Processing

1. Ran He
2. Tien-Ping Tan
3. Tongliang Liu
4. Weishi Zheng
5. Zhenan Sun
6. Bao-Gang Hu
7. Dacheng Tao
8. Xiangwei Kong

Citations in the corpus

• Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation (2014, methodology): “based on maximum correntropy criterion (MCC) [31] and”
• Robust feature learning by stacked autoencoder with maximum correntropy criterion (2014, methodology): “The correntropy induces a new metric that, as the distance between X and Y gets larger, the equivalent distance evolves from 2-norm to 1-norm and eventually to zero-norm when X and Y are far apart [11].”
• Robust Principal Component Analysis Based on Maximum Correntropy Criterion (2011, methodology): “In ITL, it has been pointed out that MCC is a local measurement whereas MSE is a global measurement [13].”
• Robust support vector machines based on the rescaled hinge loss function (2017, background): “Correntropy is an information theoretic metric [20] and has been commonly used in robust learning [21].”
• Generalized Correntropy for Robust Adaptive Filtering (2015, methodology): “The correntropy is a nonlinear and local similarity measure directly related to the probability of how similar two random variables are in a neighborhood of the joint space controlled by the kernel bandwidth, which also has its root in Renyi’s entropy (hence the name “correntropy”) [13][19].”
• Steady-State Mean-Square Error Analysis for Adaptive Filtering under the Maximum Correntropy Criterion (2014, background): “Given two random variables and , the correntropy is [3], [4] (4) where is a shift-invariantMercer kernel, and denotes the joint distribution function of .”
• Maximum Correntropy Criterion for Robust Face Recognition (2011, methodology): “Differently from SRC, which assumes that the noise term has a sparse representation, our correntropy-based sparse representation (CESR) is based on correntropy, which can efficiently cope with non-Gaussian noise and outliers [27].”
• Feature Selection Based on Structured Sparsity: A Comprehensive Study (2017, background): “Since correntropy [85] can enhance robustness, correntropy for RFS (CRFS) [86] is proposed for”
• Ieee Transactions on Pattern Analysis and Machine Intelligence Classification with Noisy Labels by Importance Reweighting (positive): “Robust surrogate loss functions, such as the Cauchy loss function [29] and correntropy (also known as the Welsch loss function), [30], [31], have been empirically proven to be robust to noise.”