To date, most linear and nonlinear Kalman filters (KFs) have been developed under the Gaussian assumption well-known minimum mean square error (MMSE) criterion. In order to improve robustness with respect impulsive (or heavy-tailed) non-Gaussian noises, maximum correntropy criterion (MCC) has recently used replace MMSE in developing several robust Kalman-type filters. deal more complicated nois...

This paper presents a new multi-sensor data fusion method based on the combination of wavelet transform (WT) and extended Kalman filter (EKF). Input data are first filtered by a wavelet transform via Daubechies wavelet “db4” functions and the filtered data are then fused based on variance weights in terms of minimum mean square error. The fused data are finally treated by extended Kalman filter...

introduction: the infiltration process is one of the most important components of the hydrologic cycle. quantifying the infiltration water into soil is of great importance in watershed management. prediction of flooding, erosion and pollutant transport all depends on the rate of runoff which is directly affected by the rate of infiltration. quantification of infiltration water into soil is also...

Sequence-discriminative training of deep neural networks (DNNs) is investigated on a 300 hour American English conversational telephone speech task. Different sequencediscriminative criteria — maximum mutual information (MMI), minimum phone error (MPE), state-level minimum Bayes risk (sMBR), and boosted MMI — are compared. Two different heuristics are investigated to improve the performance of ...

This paper studies the robustness of discriminatively trained acoustic models for large vocabulary continuous speech recognition. Popular discriminative criteria maximum mutual information (MMI), minimum phone error (MPE), and minimum phone frame error (MPFE), are used in the experiments, which include realistic mismatched conditions from Finnish Speecon corpus and English Wall Street Journal c...

In this paper, we propose minimizing the Fisher information of the error in supervised training of linear and nonlinear adaptive filters. Fisher information considers the local structure of the error probability distribution and therefore, it is expected to result in more robust solutions compared to other statistics such as minimum mean-square-error or minimum-error-entropy. A gradient-based t...