Online learning of task-driven object-based visual attention control

We propose a biologically-motivated computational model for learning task-driven and object-based visual attention control in interactive environments. In this model, top-down attention is learned interactively and is used to search for a desired object in the scene through biasing the bottom-up attention in order to form a need-based and object-driven state representation of the environment. Our model consists of three layers. First, in the early visual processing layer, most salient location of a scene is derived using the biased saliency-based bottom-up model of visual attention. Then a cognitive component in the higher visual processing layer performs an application specific operation like object recognition at the focus of attention. From this information, a state is derived in the decision making and learning layer. Top-down attention is learned by the U-TREE algorithm which successively grows an object-based binary tree. Internal nodes in this tree check the existence of a specific object in the scene by biasing the early vision and the object recognition parts. Its leaves point to states in the action value table. Motor actions are associated with the leaves. After performing a motor action, the agent receives a reinforcement signal from the critic. This signal is alternately used for modifying the tree or updating the action selection policy. The proposed model is evaluated on visual navigation tasks, where obtained results lend support to the applicability and usefulness of the developed method for robotics.     

An efficient hybrid approach based on PSO, ACO and k-means for cluster analysis

Clustering is a popular data analysis and data mining technique. A popular technique for clustering is based on k-means such that the data is partitioned into K clusters. However, the k-means algorithm highly depends on the initial state and converges to local optimum solution. This paper presents a new hybrid evolutionary algorithm to solve nonlinear partitional clustering problem. The proposed hybrid evolutionary algorithm is the combination of FAPSO (fuzzy adaptive particle swarm optimization), ACO (ant colony optimization) and k-means algorithms, called FAPSO-ACO–K, which can find better cluster partition. The performance of the proposed algorithm is evaluated through several benchmark data sets. The simulation results show that the performance of the proposed algorithm is better than other algorithms such as PSO, ACO, simulated annealing (SA), combination of PSO and SA (PSO–SA), combination of ACO and SA (ACO–SA), combination of PSO and ACO (PSO–ACO), genetic algorithm (GA), Tabu search (TS), honey bee mating optimization (HBMO) and k-means for partitional clustering problem.     

Attention control with reinforcement learning for face recognition under partial occlusion

In this paper a new method for handling occlusion in face recognition is presented. In this method the faces are partitioned into blocks and a sequential recognition structure is developed. Then, a spatial attention control strategy over the blocks is learned using reinforcement learning. The outcome of this learning is a sorted list of blocks according to their average importance in the face recognition task. In the recall mode, the sorted blocks are employed sequentially until a confident decision is made. Obtained results of various experiments on the AR face database demonstrate the superior performance of proposed method as compared with that of the holistic approach in the recognition of occluded faces.     

Genetic-based minimum classification error mapping for accurate identifying Peer-to-Peer applications in the internet traffic

In this paper, we propose a hybrid approach using genetic algorithm and neural networks to classify Peer-to-Peer (P2P) traffic in IP networks. We first compute the minimum classification error (MCE) matrix using genetic algorithm. The MCE matrix is then used during the pre-processing step to map the original dataset into a new space. The mapped data set is then fed to three different classifiers: distance-based, K-Nearest Neighbors, and neural networks classifiers. We measure three different indexes, namely mutual information, Dunn, and SD to evaluate the extent of separation of the data points before and after mapping is performed. The experimental results demonstrate that with the proposed mapping scheme we achieve, on average, 8% higher accuracy in classification of the P2P traffic compare to the previous solutions. Moreover, the genetic-based MCE matrix increases the classification accuracy more than what the basic MCE does.     

Optimal Local Basis: A Reinforcement Learning Approach for Face Recognition

This paper presents a novel learning approach for Face Recognition by introducing Optimal Local Basis. Optimal local bases are a set of basis derived by reinforcement learning to represent the face space locally. The reinforcement signal is designed to be correlated to the recognition accuracy. The optimal local bases are derived then by finding the most discriminant features for different parts of the face space, which represents either different individuals or different expressions, orientations, poses, illuminations, and other variants of the same individual. Therefore, unlike most of the existing approaches that solve the recognition problem by using a single basis for all individuals, our proposed method benefits from local information by incorporating different bases for its decision. We also introduce a novel classification scheme that uses reinforcement signal to build a similarity measure in a non-metric space. Experiments on AR, PIE, ORL and YALE databases indicate that the proposed method facilitates robust face recognition under pose, illumination and expression variations. The performance of our method is compared with that of Eigenface, Fisherface, Subclass Discriminant Analysis, and Random Subspace LDA methods as well.     

Creating corroborated crisis reports from social media data through formal concept analysis

During a crisis citizens reach for their smart phones to report, comment and explore information surrounding the crisis. These actions often involve social media and this data forms a large repository of real-time, crisis related information. Law enforcement agencies and other first responders see this information as having untapped potential. That is, it has the capacity extend their situational awareness beyond the scope of a usual command and control centre. Despite this potential, the sheer volume, the speed at which it arrives, and unstructured nature of social media means that making sense of this data is not a trivial task and one that is not yet satisfactorily solved; both in crisis management and beyond. Therefore we propose a multi-stage process to extract meaning from this data that will provide relevant and near real-time information to command and control to assist in decision support. This process begins with the capture of real-time social media data, the development of specific LEA and crisis focused taxonomies for categorisation and entity extraction, the application of formal concept analysis for aggregation and corroboration and the presentation of this data via map-based and other visualisations. We demonstrate that this novel use of formal concept analysis in combination with context-based entity extraction has the potential to inform law enforcement and/or humanitarian responders about on-going crisis events using social media data in the context of the 2015 Nepal earthquake.     

Disturbance and recovery of trunk mechanical and neuromuscular behaviours following prolonged trunk flexion: influences of duration and external load on creep-induced effects

Trunk flexion results in adverse mechanical effects on the spine and is associated with a higher incidence of low back pain. To examine the effects of creep deformation on trunk behaviours, participants were exposed to full trunk flexion in several combinations of exposure duration and external load. Trunk mechanical and neuromuscular behaviours were obtained pre- and post-exposure and during recovery using sudden perturbations. Intrinsic trunk stiffness decreased with increasing flexion duration and in the presence of the external load. Recovery of intrinsic stiffness required more time than the exposure duration and was influenced by exposure duration. Reflexive trunk responses increased immediately following exposure but recovered quickly (～2.5 min). Alterations in reflexive trunk behaviour following creep deformation exposures may not provide adequate compensation to allow for complete recovery of concurrent reductions in intrinsic stiffness, which may increase the risk of injury due to spinal instability. STATEMENT OF RELEVANCE: An increased risk of low back injury may result from flexion-induced disturbances to trunk behaviours. Such effects, however, appear to depend on the type of flexion exposure, and have implications for the design of work involving trunk flexion.     

Vehicle routing problem with uncertain demands: An advanced particle swarm algorithm

The Vehicle Routing Problem (VRP) has been thoroughly studied in the last decades. However, the main focus has been on the deterministic version where customer demands are fixed and known in advance. Uncertainty in demand has not received enough consideration. When demands are uncertain, several problems arise in the VRP. For example, there might be unmet customers’ demands, which eventually lead to profit loss. A reliable plan and set of routes, after solving the VRP, can significantly reduce the unmet demand costs, helping in obtaining customer satisfaction. This paper investigates a variant of an uncertain VRP in which the customers’ demands are supposed to be uncertain with unknown distributions. An advanced Particle Swarm Optimization (PSO) algorithm has been proposed to solve such a VRP. A novel decoding scheme has also been developed to increase the PSO efficiency. Comprehensive computational experiments, along with comparisons with other existing algorithms, have been provided to validate the proposed algorithms.