Online and Stable Learning of Analysis Operators

Sandbichler, Michael; Schnass, Karin

doi:10.1109/tsp.2018.2878540

Cited by 5 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the golden section rule, as used in [36] and [37]. Other possible directions of future work include extension of the SAD algorithm for noisy signal declipping, the employment of the dictionary pair learning (DPL) [52] for declipping, and comparisons with other analysis dictionary learning algorithms such as [53].…”

Section: Discussionmentioning

confidence: 99%

Sparse Analysis Model Based Dictionary Learning for Signal Declipping

Rencker

Dong

et al. 2021

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

Clipping is a common type of distortion in which the amplitude of a signal is truncated if it exceeds a certain threshold. Sparse representation has underpinned several algorithms developed recently for reconstruction of the original signal from clipped observations. However, these declipping algorithms are often built on a synthesis model, where the signal is represented by a dictionary weighted by sparse coding coefficients. In contrast to these works, we propose a sparse analysis-model-based declipping (SAD) method, where the declipping model is formulated on an analysis (i.e. transform) dictionary, and additional constraints characterizing the clipping process. The analysis dictionary is updated using the Analysis SimCO algorithm, and the signal is recovered by using a least-squares based method or a projected gradient descent method, incorporating the observable signal set. Numerical experiments on speech and music are used to demonstrate improved performance in signal to distortion ratio (SDR) compared to recent state-of-the-art methods including A-SPADE and ConsDL.

show abstract

Section: Discussionmentioning

confidence: 99%

Sparse Analysis Model Based Dictionary Learning for Signal Declipping

Rencker

Dong

et al. 2021

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

show abstract

“…The cosparse analysis model was recently introduced for image processing 16 . Then, several powerful recovery methods have been proposed 19–21 …”

Section: Cosparse Ce Problem Statementmentioning

confidence: 99%

“…It has been studied for cosparse signal reconstruction in image processing especially for image denoising. For example, a geometric AOL algorithm using the Riemannian gradient descent has been proposed in Hawe et al 19 ; Yaghoobi et al 20 have presented a unit norm tight frame (UNTF)‐based AOL; and four AOL algorithms: forward, sequential, implicit, and singular value AOL, have been presented in Sandbichler and Schnass 21 …”

Section: Introductionmentioning

confidence: 99%

Iterative Newton hard thresholding‐based approach for cosparse recovery of mmWave massive MIMO channels

Hajjaj

2022

Int J Communication

View full text Add to dashboard Cite

Summary Millimeter wave (mmWave) massive multiple‐input multiple‐output (MIMO) channels are cosparse in nature. These cosparse channels share common cosparsity properties, which is interesting when it comes to estimating a large number of mmWave mMIMO channels. Therefore, the channel estimation (CE) problem can be solved by exploiting the cosparsity and common cosupport properties. Specifically, the CE problem can be formulated as a cosparse signal recovery problem, which can be solved using an analysis operator (AO) and the iterative hard thresholding (IHT) algorithm. Prior analysis‐based compressive sensing (ACS) CE techniques adopt an overcomplete AO, which leads usually to a suboptimal solution that cannot guarantee CE accuracy. In this paper, we firstly propose a projected subgradient (PSG) scheme for AO learning. Then, using the previously learned AO, we propose an iterative Newton hard thresholding (INHT)‐based ACS‐CE algorithm. The effectiveness of the proposal has been proven by computational complexity analysis and numerical simulations.

show abstract

“…Films, TV dramas and other film and television art works not only penetrate into human life style and change human concept and consciousness, but also give birth to a new cultural form such as film and television literature [15]. e student-centered hybrid tutorial model can effectively integrate all kinds of teaching resources and teaching forms into the classroom [16].…”

Section: Introductionmentioning

confidence: 99%

Research on Online and Offline Mixed Teaching Practice Based on College Film and Television Literature Course

Ya-jing

2022

Scientific Programming

View full text Add to dashboard Cite

With the rapid development of new media technology, the demand for applied professional film and television talents in China’s film and television industry is getting higher and higher, especially the comprehensive practical ability. In the Internet age, information technology has been widely used. It is integrated into people’s lives with the unique characteristics of interaction and communication among all employees, and opens up a new mode of work, study and life with Internet as the medium. Under the new situation, all kinds of schools at all levels also make full and active use of the Internet for online teaching. In the past, the quality monitoring work focused on theoretical teaching. However, in the face of online and offline mixed practice teaching, there are bound to be many problems and deficiencies when using the theoretical teaching quality monitoring system. This paper mainly discusses the application of online and offline mixed tutorial mode in film and television literature course from the aspects of task-oriented teaching content, application of classroom teaching methods, diversification and dynamics of evaluation methods, etc., hoping to provide reference for college teachers who implement online and offline mixed teaching.

show abstract

Online and Stable Learning of Analysis Operators

Cited by 5 publications

References 20 publications

Sparse Analysis Model Based Dictionary Learning for Signal Declipping

Sparse Analysis Model Based Dictionary Learning for Signal Declipping

Iterative Newton hard thresholding‐based approach for cosparse recovery of mmWave massive MIMO channels

Research on Online and Offline Mixed Teaching Practice Based on College Film and Television Literature Course

Contact Info

Product

Resources

About