Upper Bound on Pattern Storage in Feedforward Networks

Narasimha, Pramod Lakshmi; Manry, M.T.; Galindo, Francisco

doi:10.1109/ijcnn.2007.4371216

Cited by 7 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since ELMs work for SLFNs, the number of hidden neurons is the only parameter that determines the networks architecture. According to [37], [63]- [66], the number of hidden neurons given in the networks is highly related to the learning capacity of the network. Varying criteria have been proposed to define the upper and lower bound of the number of hidden neurons.…”

Section: Building the Prediction Modelmentioning

confidence: 99%

Smart Information Reconstruction via Time-Space-Spectrum Continuum for Cloud Removal in Satellite Images

Chang

Bai

Chen

2015

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Cloud contamination is a big obstacle when processing satellite images retrieved from visible and infrared spectral ranges for application. Although computational techniques including interpolation and substitution have been applied to recover missing information caused by cloud contamination, these algorithms are subject to many limitations. In this paper, a novel smart information reconstruction (SMIR) method is proposed, in order to reconstruct cloud contaminated pixel values from the time-space-spectrum continuum with the aid of a machine learning tool, namely extreme learning machine (ELM). For the purpose of demonstration, the performance of SMIR is evaluated by reconstructing the missing remote sensing reflectance values derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) on board the Terra satellite over Lake Nicaragua, where is a very cloudy area year round. For comparison, the traditional backpropagation neural network algorithms will also be implemented to reconstruct the missing values. Experimental results show that the ELM outperforms the BP algorithms by an enhanced machine learning capacity with simulated memory effect embedded in MODIS due to linking the complex time-space-spectrum continuum between cloud-free and cloudy pixels. The ELM-based SMIR practice presents a correlation coefficient of 0.88 with root mean squared error of 7.4E − 04sr −1 between simulated and observed reflectance values. Finding suggests that the SMIR method is effective to reconstruct all the missing information providing visually logical and quantitatively assured images for further image processing and interpretation in environmental applications.Index Terms-Artificial neural network, cloud removal, computational intelligence, extreme learning machine, machine learning, satellite images.

show abstract

Section: Building the Prediction Modelmentioning

confidence: 99%

Smart Information Reconstruction via Time-Space-Spectrum Continuum for Cloud Removal in Satellite Images

Chang

Bai

Chen

2015

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In general in a single-layer fully connected feed-forward neural network trained with BP, the learning capacity is governed by (Huang and Babri, 1998;Narasimha et al, 2008). This rule of thumb simply states that for a given number of training samples , the number of hidden units should be more than a lower bound and still less than an upper bound that is .…”

Section: Accepted Manuscriptmentioning

confidence: 99%