Auditory BCI Research Using Spoken Digits Stimulation and Dynamic Stopping Criterion

Auditory brain-computer interfaces (BCI) provide a method of non-muscular commu-nication and control for late-stage amyotrophic lateral sclerosis (ALS) patients, who have impaired eye movements or compromised vision. In this study, random sequences of spoken digits were presented as auditory stimulation. According the protocol, the subject should pay attention to target digits and ignore non-target digits. EEG data were recorded and the components of P300 and N200 were extracted as features for pattern recognition. Fisher classifier was designed and provided likelihood estimates for the Dynamic Stopping Criterion (DSC). Dynamic data collection was controlled by a threshold of the posterior probabilities which were continually updated with each additional measurement. In addition, the experiment would be stopped and the decision was made once the probabilities were above the threshold. The results showed that this paradigm could effectively evoke the characteristic EEG, and the DSC algorithm could improve the accuracy and communication rate.

A New Stopping Criterion for Eckstein and Bertsekas’s Generalized Alternating Direction Method of Multipliers

In this paper,we propose a new stopping criterion for Eckstein and Bertsekas’s generalized alternating direction method of multipliers.The stopping criterion is easy to verify,and the computational cost is much less than the classical stopping criterion in the highly influential paper by Boyd et al.(Found Trends Mach Learn 3(1):1–122,2011).

Probability stopping criterion for analog decoding of LDPC codes

To detect uncorrectable frames and terminate the decoding procedure early, a probability stopping criterion for iterative analog decoding of low density parity check （LDPC） codes is proposed in this paper. By using probabilities of satisfied checks to detect uncorrectable frames and terminate decoding, the proposed criterion could be applied to analog decoders without much structure modifications. Simulation results show that the proposed criterion can reduce the average number of iterations and achieve a better balance in bit error ratio （BER） performance and decoding complexity than other stopping criteria using extrinsic information.

Self-Learning of Multivariate Time Series Using Perceptually Important Points

In machine learning,positive-unlabelled(PU)learning is a special case within semi-supervised learning.In positiveunlabelled learning,the training set contains some positive examples and a set of unlabelled examples from both the positive and negative classes.Positive-unlabelled learning has gained attention in many domains,especially in time-series data,in which the obtainment of labelled data is challenging.Examples which originate from the negative class are especially difficult to acquire.Self-learning is a semi-supervised method capable of PU learning in time-series data.In the self-learning approach,observations are individually added from the unlabelled data into the positive class until a stopping criterion is reached.The model is retrained after each addition with the existent labels.The main problem in self-learning is to know when to stop the learning.There are multiple,different stopping criteria in the literature,but they tend to be inaccurate or challenging to apply.This publication proposes a novel stopping criterion,which is called Peak evaluation using perceptually important points,to address this problem for time-series data.Peak evaluation using perceptually important points is exceptional,as it does not have tunable hyperparameters,which makes it easily applicable to an unsupervised setting.Simultaneously,it is flexible as it does not make any assumptions on the balance of the dataset between the positive and the negative class.

Fast Image Segmentation Based on a Two-Stage Geometrical Active Contour

A fast two-stage geometric active contour algorithm for image segmentation is developed. First, the Eikonal equation problem is quickly solved using an improved fast sweeping method, and a criterion of local minimum of area gradient (LMAG) is presented to extract the optimal arrival time. Then, the final time function is passed as an initial state to an area and length minimizing flow model, which adjusts the interface more accurately and prevents it from leaking. For object with complete and salient edge, using the first stage only is able to obtain an ideal result, and this results in a time complexity of O(M), where M is the number of points in each coordinate direction. Both stages are needed for convoluted shapes, but the computation cost can be drastically reduced. Efficiency of the algorithm is verified in segmentation experiments of real images with different feature.

Early stopping for the iterative decoding for Q-LDPC

Q-ary low-density parity-check （Q-LDPC） codes have a better performance than those of the binary low-density parity-check （B-LDPC） codes, at short and medium block lengths, but the decoder of Q-LDPC has more complexity. In this article, a new stop criterion is proposed. By analyzing the changes of the maximum posteriori probability of the variable node, the criterion decides whether the iteration of the decoder must be stopped. The simulation results show that the stop criterion can effectively reduce the computation complexity of the Q-LDPC decoder with negligible performance loss.

Multi-Objective Optimization of VBHF in Deep Drawing Based on the Improved QO-Jaya Algorithm

Blank holder force(BHF)is a crucial parameter in deep drawing,having close relation with the forming quality of sheet metal.However,there are different BHFs maintaining the best forming effect in different stages of deep drawing.The variable blank holder force(VBHF)varying with the drawing stage can overcome this problem at an extent.The optimization of VBHF is to determine the optimal BHF in every deep drawing stage.In this paper,a new heuristic optimization algorithm named Jaya is introduced to solve the optimization efficiently.An improved“Quasi-oppositional”strategy is added to Jaya algorithm for improving population diversity.Meanwhile,an innovated stop criterion is added for better convergence.Firstly,the quality evaluation criteria for wrinkling and tearing are built.Secondly,the Kriging models are developed to approximate and quantify the relation between VBHF and forming defects under random sampling.Finally,the optimization models are established and solved by the improved QO-Jaya algorithm.A VBHF optimization example of component with complicated shape and thin wall is studied to prove the effectiveness of the improved Jaya algorithm.The optimization results are compared with that obtained by other algorithms based on the TOPSIS method.