A new method for detecting P300 signals by using deep learning: Hyperparameter tuning in high-dimensional space by minimizing nonconvex error function

Abstract

Background:P300 signal detection is an essential problem in many fields of Brain-Computer Interface (BCI) systems. Although deep neural networks have almost ubiquitously used in P300 detection, in such networks, increasing the number of dimensions leads to growth ratio of saddle points to local minimums. This phenomenon results in slow convergence in deep neural network. Hyperparameter tuning is one of the approaches in deep learning, which leads to fast convergence because of its ability to find better local minimums. In this paper, a new adaptive hyperparameter tuning method is proposed to improve training of Convolutional Neural Networks (CNNs).Methods:The aim of this paper is to introduce a novel method to improve the performance of deep neural networks in P300 signal detection. To reach this purpose, the proposed method transferred the non-convex error function of CNN) into Lagranging paradigm, then, Newton and dual active set techniques are utilized for hyperparameter tuning in order to minimize error of objective function in high dimensional space of CNN.Results:The proposed method was implemented on MATLAB 2017 package and its performance was evaluated on dataset of Ecole Polytechnique Fédérale de Lausanne (EPFL) BCI group. The obtained results depicted that the proposed method detected the P300 signals with 95.34% classification accuracy in parallel with high True Positive Rate (i.e., 92.9%) and low False Positive Rate (i.e., 0.77%).Conclusions:To estimate the performance of the proposed method, the achieved results were compared with the results of Naive Hyperparameter (NHP) tuning method. The comparisons depicted the superiority of the proposed method against its alternative, in such way that the best accuracy by using the proposed method was 6.44%, better than the accuracy of the alternative method.

Current Status, Challenges, and Possible Solutions of EEG-Based Brain-Computer Interface: A Comprehensive Review

“…In BCI research, CNN is assumed to capture the distinctive dependencies amongst the patterns associated with different brain signals (Lotte et al, 2018). Recently, a considerable amount of studies (Tang et al, 2017;Aznan et al, 2018;Dose et al, 2018;El-Fiqi et al, 2018;Shojaedini et al, 2018;Wang et al, 2018;Waytowich et al, 2018;Amber et al, 2019;Amin et al, 2019;Nguyen and Chung, 2019;Olivas-Padilla and Chacon-Murguia, 2019;Tayeb et al, 2019;Xu et al, 2019) on the employment of CNN architecture in EEG-based BCI systems have been published. In Olivas-Padilla and Chacon-Murguia ( 2019), the classification of multiple MI using CNN was explored, with the features being extracted by a variety of Discriminative Filter Bank Common Spatial Patterns (DFBCSP).…”

“…With regard to the use of CNN on P300, Amber et al (2019) presented a lie detection system from the P300 signals with an accuracy of 99.6%. In addition, a new adaptive hyperparameter-tuning method is proposed in Shojaedini et al (2018) to improve the training of CNN in P300 signal detection. It was established from the study that the proposed method is able to improve the classification accuracy by 6.44% against the conventional Naive hyperparameter tuning method.…”

An Integrative Computational Approach for the Prediction of Human-Plasmodium Protein-Protein Interactions

“…K is fixed at 10, which means that the training dataset is divided into 10 equal parts and the process will run 10 times, each with a different holdout set. This allows us to keep our test set as an unseen dataset for selecting the final tuned model [50]. To evaluate the performances of the studied classifiers, we estimated the five measures below:…”

Thought-actuated wheelchair navigation with communication assistance using statistical cross-correlation-based features and extreme learning machine

“…Cross-correlation-based techniques have been proposed in this analysis to derive the features from the frequency-band signals at every electrode channel. [ 33 ] As many classifiers such as backpropagation neural network,[ 34 ] SVM,[ 35 ] and LDA[ 36 ] have also shown greater efficiency in recognition, the proposed extreme learning machine (ELM) by Huang et al ., 2012, is indeed an effective tuning-free algorithm for training a feature set that employs simply single-hidden layer in feed-forward neural network (SLFNs). [ 37 38 ] The emergence of ELM in the artificial neural nets allows reduced time for training the network models relative to the artificial neural network, which has also been employed in other areas of research, particularly in BCI.…”