Journal: Frontiers in Human Neuroscience

Article Title: ERP Subsequent Memory Effects Differ between Inter-Item and Unitization Encoding Tasks

doi: 10.3389/fnhum.2017.00030

Figure Lengend Snippet: Experimental design: Trial structure at encoding and recognition. (A) Participants were randomly assigned to either the definition- or the sentence encoding condition. Encoding event-related potential (ERPs) were time locked to the onset of the word pair. (B) The test structure was identical for participants in both conditions. In both the encoding and test phase, the inter-trial interval (from the offset of the rating screen to the onset of the next sentence/definition or the next word pair, respectively) was 2000 ms. Example stimuli have been translated from German to English (The German word pair is “Familien, Bikini”; the first word is in the German version in plural form to allow for the pair to be a legal compound word. See main text for details).

Article Snippet: To this end, in an additional analysis step the subject averages for all electrodes and for subsequent hits and subsequent misses were submitted to a spatio-temporal PCA ( ) using the Matlab-based ERP PCA toolkit ( ).

Techniques:

Journal: Frontiers in Human Neuroscience

Article Title: ERP Subsequent Memory Effects Differ between Inter-Item and Unitization Encoding Tasks

doi: 10.3389/fnhum.2017.00030

Figure Lengend Snippet: Pearson’s correlation coefficients between Pr scores and event-related potential (ERP) subsequent memory effects (SMEs).

Article Snippet: To this end, in an additional analysis step the subject averages for all electrodes and for subsequent hits and subsequent misses were submitted to a spatio-temporal PCA ( ) using the Matlab-based ERP PCA toolkit ( ).

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: Ideal LSTM architectures with the Denoising Stage.

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: Denoising Autoencoder (DAE) architectures. x ∈ R m × 1 corresponds to the input data. w ∈ R m × 1 corresponds to the noise corrupting the input data vector. z ∈ R k × 1 vector corresponds to the compressed data which is represented in a space with k dimensions. Finally, x ∈ R m × 1 is the denoised data vector. Nodes in red correspond to the DAE latent space.

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques: Plasmid Preparation

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: Performance of the different denoising approaches. The denoising metrics have been computed adopting the average of the denoising process of each input variable. The best denoising approach is in bold.

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: S O , 4 denoising process. The measured signal is depicted in blue, the denoised and real ones are shown in orange and yellow, respectively.

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: Stability analysis for the different denoising approaches.

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Denoising Autoencoders and LSTM-Based Artificial Neural Networks Data Processing for Its Application to Internal Model Control in Industrial Environments—The Wastewater Treatment Plant Control Case

doi: 10.3390/s20133743

Figure Lengend Snippet: Tracking process of the S O , 5 [ n ] concentration. The three different denoising approaches have been considered, however, the one offering the best tracking is the Multilayer Perceptron-Sliding Window (MLP-SW).

Article Snippet: Matlab has been considered to design the PCA-based denoising approach.

Techniques: Concentration Assay