The process of recording Electroencephalography (EEG) signals is onerous and requires massive storage to store signals at an applicable frequency rate. In this work, we propose the Event-Related Potential Encoder Network (ERPENet); a multi-task autoencoder-based model, that can be applied to any ERP-related tasks. The strength of ERPENet lies in its capability to handle various kinds of ERP datasets and its robustness across multiple recording setups, enabling joint training across datasets. ERPENet incorporates Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM), in an autoencoder setup, which tries to simultaneously compress the input EEG signal and extract related P300 features into a latent vector. Here, we can infer the process for generating the latent vector as universal joint feature extraction. The network also includes a classification part for attended and unattended events classification as an auxiliary task. We experimented on six different P300 datasets. The results show that the latent vector exhibits better compression capability than the previous state-of-the-art semi-supervised autoencoder model. For attended and unattended events classification, pre-trained weights are adopted as initial weights and tested on unseen P300 datasets to evaluate the adaptability of the model, which shortens the training process as compared to using random Xavier weight initialization. At the compression rate of 6.84, the classification accuracy outperforms conventional P300 classification models: XdawnLDA, DeepConvNet, and EEGNet achieving 79.37% -88.52% classification accuracy depending on the dataset. INDEX TERMS Electroencephalography, P300, Deep learning, Pre-trained model, Spatiotemporal neural networks, multi-task autoencoder arXiv:1808.06541v2 [eess.SP]
The oxidation of CO by NO over metal-organic framework (MOF) M(btc) (M = Fe, Cr, Co, Ni, Cu, and Zn) catalysts that contain coordinatively unsaturated sites has been investigated by means of density functional theory calculations. The reaction proceeds in two steps. First, the N-O bond of NO is broken to form a metal oxo intermediate. Second, a CO molecule reacts with the oxygen atom of the metal oxo site, forming one C-O bond of CO. The first step is a rate-determining step for both Cu(btc) and Fe(btc), where it requires the highest activation energy (67.3 and 19.6 kcal/mol, respectively). The lower value for the iron compound compared to the copper one can be explained by the larger amount of electron density transferred from the catalytic site to the antibonding of NO molecules. This, in turn, is due to the smaller gap between the highest occupied molecular orbital (HOMO) of the MOF and the lowest unoccupied molecular orbital (LUMO) of NO for Fe(btc) compared to Cu(btc). The results indicate the important role of charge transfer for the N-O bond breaking in NO. We computationally screened other MOF M(btc) (M = Cr, Fe, Co, Ni, Cu, and Zn) compounds in this respect and show some relationships between the activation energy and orbital properties like HOMO energies and the spin densities of the metals at the active sites of the MOFs.
First-principles gradient-corrected density functional theory electronic structure calculations of the haptotropic rearrangement of a Cr(CO)(3) unit on naphthalene and phenanthrene derivatives are reported. Coupled-cluster calibration studies of Cr(CO)(3) complexes with benzene and naphthalene derivatives confirm the accuracy of the applied Becke exchange and Perdew correlation functionals. Characteristic points on the energy hypersurface (reactants, products, intermediates, and transition states) were located for various substituents on the aromatic skeleton. It is argued that a -OH/-O(-) substituent may provide a means to steer the haptotropic shift depending on the pH value, i.e., to construct a molecular switch. In addition, the mechanism of the [3 + 2 + 1] benzannulation of chromium pentacarbonyl naphthylcarbene complexes with alkynes was investigated, and the preference of the angular benzannulation leading to phenanthrene complexes of Cr(CO)(3) over the linear benzannulation leading to corresponding anthracene complexes is explained.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.