We propose a new source connectivity method by focusing on estimating time courses of the regions of interest (ROIs). To this aim, it is necessary to consider the strong inherent non-stationary behavior of neural activity. We develop an iterative dynamic approach to extract a single time course for each ROI encoding the temporal non-stationary features. The proposed approach explicitly includes dynamic constraints by taking into account the evolution of the sources activities for further dynamic connectivity analysis. We simulated an epileptic network with a non-stationary structure; accordingly, EEG source reconstruction using LORETA is performed. Using the reconstructed sources, the spatially compact ROIs are selected. Then, a single time course encoding the temporal non-stationarity is extracted for each ROI. An adaptive directed transfer function (ADTF) is applied to measure the information flow of underlying brain networks. Obtained results demonstrate that the contributed approach is more efficient to estimate the ROI time series and ROI to ROI information flow in comparison with existing methods. Our work is validated in three drug-resistance epilepsy patients. The proposed ROI time series estimation directly affects the quality of connectivity analysis, leading to the best possible seizure onset zone (SOZ) localization verified by electrocorticography and post-operational results.

There is a variety of fuzzy classifiers, one of which is Adaptive Neuro-Fuzzy Inference system (ANFIS) classifier. One of the main challenges in designing such data classifiers is selection of effective and appropriate type and location of membership functions and its training method to reduce the classification error. In this paper, a new technique (based on intelligent methods) is presented and implemented to select and locate the membership functions and simultaneous training using a new method based on Inclined Planes System Optimization (IPO) to minimize errors of an ANFIS classifier for the first time. The presented method is evaluated for classification of data sets with different reference classes and different length feature vectors, which have acceptable complexity. According to the results of the research, the presented method has a higher level of accuracy and efficiency in selecting the type and location of membership functions (based on intelligent methods) and simultaneous training with IPO, compared to other methods, such as particle swarm optimization, genetic algorithm, differential evolution, and ACOR algorithms.

This paper outlines a neural model inspired by the dorsal stream of the visual system for motion recognition. Two areas are considered: the primary visual area (V1) and the middle temporal area (MT). In model area, V1 neurons are organized to detect eight local motion directions. MT is modelled using classical receptive field (CRF), where the cells respond to wide-field motion. The biological motion can be identified through spatial motion dynamics for the limbs and body. In this article, we propose spatio-temporal sampling detectors, where a set of circular masks over motion scenario are utilized to detect the motion dynamics. Two alternative mechanisms, Max-pooling and Sum-pooling, are used to extracting spatio-temporal descriptors from motion energy occupied by the circular masks. To improve the classification results, centroid kinematics is added to the feature vectors, where this feature contributes substantially to characterizing the motion pattern of an action. We evaluate our model by using two challenging datasets: the Weizmann biological action dataset and the KTH biological motion dataset. Our results reflect the potential of spatio-temporal sampling detectors in describing the biological motion of body and limbs using only short video frames (snippets). In addition, the centroid kinematic feature improves the recognition rate and refines the action classification.

Autonomous navigation in dynamic environment is aprerequisite of the mobile robot to perform tasks, and numerous approaches have been presented, including the supervised learning. Using supervised learning in robot navigation might meet problems, such as inconsistent and noisy data, and high error in training data. Inspired by the advantages of the reinforcement learning, such as no need for desired outputs, many researchers have applied reinforcement learning to robot navigation. This paper presents anovel method to address the robot navigation in different settings, through integrating supervised learning and analogical reinforcement learning into amotivated developmental network. We focus on the effect of the new learning rate on the robot navigation behavior. Experimentally, we show that the effect of internal neurons on the learning rate allows the agent to approach the target and avoid the obstacle as compounding effects of sequential states in static, dynamic, and complex environments. Further, we compare the performance between the emergent developmental network system and asymbolic system, as well as other four reinforcement learning algorithms. These experiments indicate that the reinforcement learning is beneficial for developing desirable behaviors in this set of robot navigation– staying statistically close to its target and away from obstacle.

This paper presents alternative continuous- and discrete-time neural networks for image restoration in real time by introducing new vectors and transforming its optimization conditions into a system of double projection equations. The proposed neural networks are shown to be stable in the sense of Lyapunov and convergent for any starting point. Compared with the existing neural networks for image restoration, the proposed models have the least neurons, a one-layer structure and the faster convergence, and is suitable to parallel implementation. The validity and transient behaviour of the proposed neural network is demonstrated by numerical examples.

This paper presents a method based on neural networks to solve fractional infinite-horizon optimal control problems s(FIHOCP)s, where the dynamic control system depends on Caputo fractional derivatives. First, with the help of an approximation, the Caputo derivative is replaced to integer-order derivative. Using a suitable change of variable, the IHOCP is transformed into a finite-horizon one. According to the Pontryagin minimum principle (PMP) for optimal control problems and by constructing an error function, an unconstrained minimization problem is defined. In the optimization problem, the trial solutions are used for state, costate and control functions where these trial solutions are constructed by using two-layered perceptron neural network. Two numerical results are introduced to explain our main results.

This paper solved fractional order optimal control problems, in which the dynamic control system involves integer and fractional order derivatives with equality and inequality constraints. According to the Pontryagin minimum principle (PMP) for fractional optimal control problem (FOCP) with fractional derivative in the Caputo sense and by constructing a suitable error function, an unconstrained minimization problem is defined. In the optimization problem, trial solutions are used for the states, Lagrange multipliers and control functions where these trial solutions are constructed by fractional power series neural network models. An error function is then minimized using a numerical optimization scheme where weight parameters (or coefficients of the series) and biases associated with all neurons are unknown. Some computational simulations are discussed in details.