Nonlinear model predictive control with relevance vector regression and particle swarm optimization

In this paper, a nonlinear model predictive control strategy which utilizes a probabilistic sparse kernel learning technique called relevance vector regression （RVR） and particle swarm optimization with controllable random exploration velocity （PSO-CREV） is applied to a catalytic continuous stirred tank reactor （CSTR） process. An accurate reliable nonlinear model is first identified by RVR with a radial basis function （RBF） kernel and then the optimization of control sequence is speeded up by PSO-CREV. Additional stochastic behavior in PSO-CREV is omitted for faster convergence of nonlinear optimization. An improved system performance is guaranteed by an accurate sparse predictive model and an efficient and fast optimization algorithm. To compare the performance, model predictive control （MPC） using a deterministic sparse kernel learning technique called Least squares support vector machines （LS-SVM） regression is done on a CSTR. Relevance vector regression shows improved tracking performance with very less computation time which is much essential for real time control.

Decoding fear of negative evaluation from brain morphology:A machine-learning study on structural neuroimaging data

Background:Fear of negative evaluation(FNE),referring to negative expectation and feelings toward other people’s social evaluation,is closely associated with social anxiety that plays an important role in our social life.Exploring the neural markers of FNE may be of theoretical and practical significance to psychiatry research(e.g.,studies on social anxiety).Methods:To search for potentially relevant biomarkers of FNE in human brain,the current study applied multivariate relevance vector regression,a machine-learning and data-driven approach,on brain morphological features(e.g.,cortical thickness)derived from structural imaging data;further,we used these features as indexes to predict self-reported FNE score in each participant.Results:Our results confirm the predictive power of multiple brain regions,including those engaged in negative emotional experience(e.g.,amygdala,insula),regulation and inhibition of emotional feeling(e.g.,frontal gyrus,anterior cingulate gyrus),and encoding and retrieval of emotional memory(e.g.,posterior cingulate cortex,parahippocampal gyrus).Conclusions:The current findings suggest that anxiety represents a complicated construct that engages multiple brain systems,from primitive subcortical mechanisms to sophisticated cortical processes.

A RVR-based Method for Bias Field Estimation in Brain Magnetic Resonance Images Segmentation

This paper presents a relevance vector regression(RVR) based on parametric approach to the bias field estimation in brain magnetic resonance(MR) image segmentation. Segmentation is a very important and challenging task in brain analysis,while the bias field existed in the images can significantly deteriorate the performance.Most of current parametric bias field correction techniques use a pre-set linear combination of low degree basis functions, the coefficients and the basis function types of which completely determine the field. The proposed RVR method can automatically determine the best combination for the bias field, resulting in a good segmentation in the presence of noise by combining with spatial constrained fuzzy C-means(SCFCM)segmentation. Experiments on simulated T1 images show the efficiency.