Untitled

Bertels, Koen; Neuberg, Luc; Vassiliadis, S.; Pechanek, D. G.

doi:10.1023/a:1011045100817

Cited by 13 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many techniques have been developed for prediction of time series, including autoregressive moving average (Box et al 1994), nearest neighborhood prediction (Garcia and Almeida 2005), adaptive filters (Hartmann 2010), Kalman filters (Silva 2010), radial basis function (Leung et al 2001, Zou et al 2003, Song et al 2005, wavelet decomposition (Wu 2010), artificial neural networks (Bertels et al 2001, Han et al 2004, Hastie et al 2001, Haykin 1999, Rodrigues 2010, principal component analysis (Petrolis et al 2010), and many others (Kantz and Schreiber 2004, Li et al 2010, Langley et al 2010. Our preliminary studies using a few different techniques indicate that neural networks are more robust compared to others.…”

Section: Methodsmentioning

confidence: 99%

Reconstruction of physiological signals using iterative retraining and accumulated averaging of neural network models

et al. 2011

View full text Add to dashboard Cite

Real-time monitoring of vital physiological signals is of significant clinical relevance. Disruptions in the signals are frequently encountered and make it difficult for precise diagnosis. Thus, the ability to accurately predict/recover the lost signals could greatly impact medical research and application. We have developed new techniques of signal reconstructions based on iterative retraining and accumulated averaging of neural networks. The effectiveness and robustness of these techniques are demonstrated using data records from the Computing in Cardiology/PhysioNet Challenge 2010. The average correlation coefficient between prediction and target for 100 records of various target signals is about 0.9. We have also explored influences of a few important parameters on the accuracy of reconstructions. The developed techniques may be used to detect changes in patient state and to recognize intervals of signal corruption.

show abstract

Section: Methodsmentioning

confidence: 99%

Reconstruction of physiological signals using iterative retraining and accumulated averaging of neural network models

et al. 2011

View full text Add to dashboard Cite

show abstract

“…For the hidden nodes, the error can be computed in the following way: (5) where ␦ pk W kj represents the error of the connected neurons of the above layer, multiplied by the corresponding weights, which is propagated throughout the net. For exactly the same reasons as mentioned above, the first derivative of the transfer function is included.…”

Section: Figurementioning

confidence: 99%

“…For a detailed discussion of neural networks, we refer to Ref. [2], which previously appeared in Complexity, and for an extensive discussion of the results reported here we refer to Refs.[3], [4], and [5]. The following restrictions apply to the research reported on in this article.…”

mentioning

confidence: 99%

A look inside the learning process of neural networks

et al. 2000

Self Cite

View full text Add to dashboard Cite

ne of the best-known results of the sciences of complexity is that complex systems learn on the edge of chaos, by which is meant that both chaotic and orderly states coexist and that the system remains close to this borderline and may switch from one state to the other. In this article, we take a look inside the learning process of neural networks, and we more specifically focus on the role of chaos for learning a nonlinearly separable function, the XOR Boolean function. It is clear that neural networks are complex adaptive systems (CAS) in the sense of Casti's definition of CAS [1]. The neurones are individual components that can make some kind of computation. These neurones interact as they send the result of their computation to neighboring neurones, and they only use local information as they do not receive information from all the neurones in the network. For a detailed discussion of neural networks, we refer to Ref. [2], which previously appeared in Complexity, and for an extensive discussion of the results reported here we refer to Refs.[3], [4], and [5]. The following restrictions apply to the research reported on in this article. We only look at a relatively simple problem. Even though it could be argued that a simple problem such as the XOR problem is not representative of more realistic problems, the simplicity of the network allows us to better focus on the phenomenon we want to study, namely the occurrence of chaos during the learning process. We also restrict ourselves to multilayer neural networks that are trained, using the backpropagation algorithm. Similar analyses for other learning algorithms could be the subject of further research.The article is structured as follows. We first briefly introduce the basic equations of the backpropagation algorithm (BPA). We then show that chaos occurs during the learning process and investigate in more detail the role of two parameters, the learning rate and the temperature. We finally address the question whether chaos increases learning speed and could therefore be considered necessary or beneficial for efficient learning. ALGORITHMIC BASICSWe will not explain the entire algorithm here as it has already been discussed previously, but the remainder of the article centers around the so-called Delta rule. As we know, backpropagation neural networks belong to the category of supervised learning neural networks, implying that it is known what output the network has to produce for a given set of inputs. The following equations allow us to compute the output of any node (hidden or output

show abstract

Wandel und Kontinuität von Wissenschaft durch KI. Zur aktuellen Veränderung des Wissenschafts- und Technikverständnisses

Schmidt

2021

Ethics of Science and Technology Assessment

View full text Add to dashboard Cite

show abstract

Untitled

Cited by 13 publications

References 11 publications

Reconstruction of physiological signals using iterative retraining and accumulated averaging of neural network models

Reconstruction of physiological signals using iterative retraining and accumulated averaging of neural network models

A look inside the learning process of neural networks

Wandel und Kontinuität von Wissenschaft durch KI. Zur aktuellen Veränderung des Wissenschafts- und Technikverständnisses

Contact Info

Product

Resources

About