Journal articles on the topic 'Escaping local minima'

Inspired by the reflective and deliberative control mechanisms used in cognitive architectures such as SOAR and Sigma, we propose an alternative decision mechanism driven by architectural appraisals allowing robots to overcome impasses. The presented work builds on and improves on our previous work on a generally applicable decision mechanism with roots in the Standard Model of the Mind and the Generalized Cognitive Hour-glass Model. The proposed decision mechanism provides automatic context-dependent switching between exploration-oriented, goal-oriented, and backtracking behavior, allowing a robot to overcome impasses. A simulation study of two applications utilizing the proposed decision mechanism is presented demonstrating the applicability of the proposed decision mechanism.

Abstract Artificial Potential Field (APF) is one of path planning strategies which offers a relatively low cost on computational. It has been implemented in many real-time applications. Unfortunately, it is quite susceptible to local minima problems. In this paper, we offer a scheme for escaping local minima by using vortex field. Based on the results and simulation, it can be verified that the vortex field prevails to navigate the path while escaping local minima.

The bacterial foraging optimization (BFO) algorithm successfully searches for an optimal path from start to finish in the presence of obstacles over a flat surface map. However, the algorithm suffers from getting stuck in the local minima whenever non-circular obstacles are encountered. The retrieval from the local minima is crucial, as otherwise, it can cause the failure of the whole task. This research proposes an improved version of BFO called robust bacterial foraging (RBF), which can effectively avoid obstacles, both of circular and non-circular shape, without falling into the local minima. The virtual obstacles are generated in the local minima, causing the robot to retract and regenerate a safe path. The proposed method is easily extendable to multiple robots that can coordinate with each other. The information related to the virtual obstacles is shared with the whole swarm, so that they can escape the same local minima to save time and energy. To test the effectiveness of the proposed algorithm, a comparison is made against the existing BFO algorithm. Through the results, it was witnessed that the proposed approach successfully recovered from the local minima, whereas the BFO got stuck.

A method of supervised learning for multilayer artificial neural networks to escape local minima is proposed. The learning model has two phases: a backpropagation phase and a gradient ascent phase. The backpropagation phase performs steepest descent on a surface in weight space whose height at any point in weight space is equal to an error measure, and it finds a set of weights minimizing this error measure. When the backpropagation gets stuck in local minima, the gradient ascent phase attempts to fill up the valley by modifying gain parameters in a gradient ascent direction of the error measure. The two phases are repeated until the network gets out of local minima. The algorithm has been tested on benchmark problems, such as exclusive-or (XOR), parity, alphabetic characters learning, Arabic numerals with a noise recognition problem, and a realistic real-world problem: classification of radar returns from the ionosphere. For all of these problems, the systems are shown to be capable of escaping from the backpropagation local minima and converge faster when using the new proposed method than using the simulated annealing techniques.

The purpose of this paper is to present a new hybrid Elastic Net (EN) algorithm, by integrating the ideas of the Self Organization Map (SOM) and the strategy of the gradient ascent into the EN algorithm. The new hybrid algorithm has two phases: an EN phase based on SOM and a gradient ascent phase. We acquired the EN phase based on SOM by analyzing the weight between a city and its converging and non-converging nodes at the limit when the EN algorithm produces a tour. Once the EN phase based on SOM stuck in local minima, the gradient ascent algorithm attempts to fill up the valley by modifying parameters in a gradient ascent direction of the energy function. These two phases are repeated until the EN gets out of local minima and produces the short or better tour through cities. We test the algorithm on a set of TSP. For all instances, the algorithm is showed to be capable of escaping from the EN local minima and producing more meaningful tour than the EN.

Abstract Artificial Potential Filed (APF) is the most well-known method that is used in mobile robot path planning, however, the shortcoming is that the local minima. To overcome this issue, we present a deadlock free APF based path planning algorithm for mobile robot navigation. The Proposed-APF (P-APF) algorithm searches the goal point in unknown 2D environments. This method is capable of escaping from deadlock and non-reachability problems of mobile robot navigation. In this method, the effective front-face obstacle information associated with the velocity direction is used to modify the Traditional APF (T-APF) algorithm. This modification solves the deadlock problem that the T-APF algorithm often converges to local minima. The proposed algorithm is explained in details and to show the effectiveness of the proposed approach, the simulation experiments were carried out in the MATLAB environment. Furthermore, the numerical analysis of the proposed approach is given to prove a deadlock free motion of the mobile robot.

In this paper, the authors propose an adaptive Elastic Net method for edge linking of images. Edge linking is a fundamental computer-vision task, which is a constrained optimization problem. In the proposed method, an adaptive dynamic parameter strategy and a stochastic noise strategy are introduced into the Elastic Net, which enables the network to have superior ability for escaping from local minima and converge sooner to optimal or near-optimal solutions. Simulations confirm that the proposed method could produce more meaningful contours than the original Elastic Net in shorter time.

To solve the real-time path planning of multi-robots in complex environments, a new immune planning algorithm incorporating a specific immune mechanism is presented. In the immune planning algorithm incorporating a specific immune mechanism, a new coding format for an antibody is first defined according to the impact of the obstacle distribution on the obstacle avoidance behaviors of multi-robots. Then, a new robot immune dynamic model for antibody selection is designed in terms of different impacts of obstacles and targets on robot behaviors. Finally, aiming at the local minimum problem in complex environments and inspired by the specific immune mechanism, a series of appropriate avoidance behaviors are selected through the calculation of a specific immune mechanism to help robots walk out of local minima. In addition, to solve deadlock situations, a learning strategy for the antibody concentration is presented. Compared with four related immune planning algorithms—an improved artificial potential field, a rapidly exploring random tree algorithm, a D* algorithm and a A* algorithm—the simulation results in four static environments show that the paths planned by immune planning algorithm incorporating a specific immune mechanism are the shortest and the path smoothness is generally the highest, which shows its strong planning capability in multi-obstacle environments. The simulation result in a dynamic environment with local minima shows that the immune planning algorithm incorporating a specific immune mechanism has strong planning ability in dynamic obstacle avoidance and in escaping from local minima. Additionally, an experiment in a multi-robot environment shows that two robots can not only avoid static obstacles but also avoid dynamic obstacles, which further supports the validity of the proposed immune planning algorithm incorporating a specific immune mechanism for multi-robots in real environments.

An extended pattern search algorithm is introduced for efficient component layout optimization. The algorithm is applicable to general layout problems, where component geometry can be arbitrary, design goals can be multiple and spatial constraint satisfactions can be of different types. Extensions to pattern search are introduced to help the algorithm to converge to optimal solutions by escaping inferior local minima. The performance on all of the test problems shows that the algorithm runs one-to-two orders of magnitude faster than a robust simulated annealing-based algorithm for results with the same quality. The algorithm is further extended to solve a concurrent layout and routing problem, which demonstrates the ability of the algorithm to apply new pattern strategies in search and to include different objective functions in optimization. [S1050-0472(00)01901-2]

Abstract This paper presents the identification of the ThermoElectric Cooler (TEC) plant using a hybrid method of Multi-Verse Optimizer with Sine Cosine Algorithm (hMVOSCA) based on continuous-time Hammerstein model. These modifications are mainly for escaping from local minima and for making the balance between exploration and exploitation. In the Hammerstein model identification a continuous-time linear system is used and the hMVOSCA based method is used to tune the coefficients of both the Hammerstein model subsystems (linear and nonlinear) such that the error between the estimated output and the actual output is reduced. The efficiency of the proposed method is evaluated based on the convergence curve, parameter estimation error, bode plot, function plot, and Wilcoxon’s rank test. The experimental findings show that the hMVOSCA can produce a Hammerstein system that generates an estimated output like the actual TEC output. Moreover, the identified outputs also show that the hMVOSCA outperforms other popular metaheuristic algorithms.

One of the most widely used algorithms to solve clustering problems is the K-means. Despite of the algorithm's timely performance to find a fairly good solution, it shows some drawbacks like its dependence on initial conditions and trapping in local minima. This paper proposes a novel hybrid algorithm, comprised of K-means and a variation operator inspired by mutation in evolutionary algorithms, called Noisy K-means Algorithm (NKA). Previous research used K-means as one of the genetic operators in Genetic Algorithms. However, the proposed NKA is a kind of individual based algorithm that combines advantages of both K-means and mutation. As a result, proposed NKA algorithm has the advantage of faster convergence time, while escaping from local optima. In this algorithm, a probability function is utilized which adaptively tunes the rate of mutation. Furthermore, a special mutation operator is used to guide the search process according to the algorithm performance. Finally, the proposed algorithm is compared with the classical K-means, SOM Neural Network, Tabu Search and Genetic Algorithm in a given set of data. Simulation results statistically demonstrate that NKA out-performs all others and it is prominently prone to real time clustering.

Despite of the many successful applications of backpropagation for training multi–layer neural networks, it has many drawbacks. For complex problems it may require a long time to train the networks, and it may not train at all. Long training time can be the result of the non-optimal parameters. It is not easy to choose appropriate value of the parameters for a particular problem. In this paper, by interconnection of fixed structure learning automata (FSLA) to the feedforward neural networks, we apply learning automata (LA) scheme for adjusting these parameters based on the observation of random response of neural networks. The main motivation in using learning automata as an adaptation algorithm is to use its capability of global optimization when dealing with multi-modal surface. The feasibility of proposed method is shown through simulations on three learning problems: exclusive-or, encoding problem, and digit recognition. The simulation results show that the adaptation of these parameters using this method not only increases the convergence rate of learning but it increases the likelihood of escaping from the local minima.

High education is an important and critical part of education all over the world. In last year, the world has been turned increasingly to online education due to the outbreak of the Covid-19 pandemic; therefore, improving this education system became an urgent matter. Online learning systems are a primal environment for acquiring educational data which can be from different sources, especially academic institutions. These data can be mainly used to analyze and extract utilizable information to help in understanding university students’ performance and identifying factors that affect it. To extract some meaningful information from these large volumes of data, academic organizations must mine the data with high accuracy. In this work, three different real datasets were selected, pre-processed, cleaned, and filtered for applying support vector machine (SVM) with multilayer perceptron kernel (MLP kernel) and optimize its parameters using simulated annealing (SA) algorithm to improve the objective function value. While examining the search space, SA has the advantage of escaping from local minima since it offers the chance for accepting the worse neighbor as a solution in a controlled manner. The results show that the designed system can determine the best SVM parameters using SA and therefore presents better model evaluation.

Between 1998 and 2004, the planning community has seen vast progress in terms of the sizes of benchmark examples that domain-independent planners can tackle successfully. The key technique behind this progress is the use of heuristic functions based on relaxing the planning task at hand, where the relaxation is to assume that all delete lists are empty. The unprecedented success of such methods, in many commonly used benchmark examples, calls for an understanding of what classes of domains these methods are well suited for. In the investigation at hand, we derive a formal background to such an understanding. We perform a case study covering a range of 30 commonly used STRIPS and ADL benchmark domains, including all examples used in the first four international planning competitions. We *prove* connections between domain structure and local search topology -- heuristic cost surface properties -- under an idealized version of the heuristic functions used in modern planners. The idealized heuristic function is called h^+, and differs from the practically used functions in that it returns the length of an *optimal* relaxed plan, which is NP-hard to compute. We identify several key characteristics of the topology under h^+, concerning the existence/non-existence of unrecognized dead ends, as well as the existence/non-existence of constant upper bounds on the difficulty of escaping local minima and benches. These distinctions divide the (set of all) planning domains into a taxonomy of classes of varying h^+ topology. As it turns out, many of the 30 investigated domains lie in classes with a relatively easy topology. Most particularly, 12 of the domains lie in classes where FF's search algorithm, provided with h^+, is a polynomial solving mechanism. We also present results relating h^+ to its approximation as implemented in FF. The behavior regarding dead ends is provably the same. We summarize the results of an empirical investigation showing that, in many domains, the topological qualities of h^+ are largely inherited by the approximation. The overall investigation gives a rare example of a successful analysis of the connections between typical-case problem structure, and search performance. The theoretical investigation also gives hints on how the topological phenomena might be automatically recognizable by domain analysis techniques. We outline some preliminary steps we made into that direction.

In this paper, the one-diode model of a photovoltaic PV solar cell (PVSC) is estimated for an experimental characteristic curves data by using a recently proposed version of the Particle Swarm Optimization (PSO) algorithm, which is known as the Autonomous Groups Particles Swarm Optimization (PSOAG). This meta-heuristic algorithm is used to identify the model of the PVSC. The PSOAG divides the particles into groups and then, uses different functions to tune the social and cognitive parameters of these groups. This is done to show the individuals’ diversity inside the swarm. Although, these individuals do their duties as part of the society, they are not similar in terms of intelligence and ability. By using these groups, the performance of the PSO is improved in terms of convergence rate and escaping the local minima/maxima. Six versions of PSOAG algorithms were developed in this work. Therefore, nine versions of PSOAG, including these six algorithms and three newly developed PSOAG reported previously, will be used in this research to cover more social’s behaviors. The results are compared to the original PSO and other versions of PSO like conventional and Asymmetric Time-varying Accelerated Coefficient PSOs, and the improved PSO. The result shows that the proposed methods improve the performance by up to 14% in terms of root mean squared error and maximum absolute error, and by up to 20% in term of convergence rate, when these were compared to the best results obtained from the other algorithms.

Purpose The purpose of this paper is to propose an enhanced elephant herding optimization (EEHO) algorithm by improving the exploration phase to overcome the fast-unjustified convergence toward the origin of the native EHO. The exploration and exploitation of the proposed EEHO are achieved by updating both clan and separation operators. Design/methodology/approach The original EHO shows fast unjustified convergence toward the origin specifically, a constant function is used as a benchmark for inspecting the biased convergence of evolutionary algorithms. Furthermore, the star discrepancy measure is adopted to quantify the quality of the exploration phase of evolutionary algorithms in general. Findings In experiments, EEHO has shown a better performance of convergence rate compared with the original EHO. Reasons behind this performance are: EEHO proposes a more exploitative search method than the one used in EHO and the balanced control of exploration and exploitation based on fixing clan updating operator and separating operator. Operator γ is added to EEHO assists to escape from local optima, which commonly exist in the search space. The proposed EEHO controls the convergence rate and the random walk independently. Eventually, the quantitative and qualitative results revealed that the proposed EEHO outperforms the original EHO. Research limitations/implications Therefore, the pros and cons are reported as follows: pros of EEHO compared to EHO – 1) unbiased exploration of the whole search space thanks to the proposed update operator that fixed the unjustified convergence of the EHO toward the origin and the proposed separating operator that fixed the tendency of EHO to introduce new elephants at the boundary of the search space; and 2) the ability to control exploration–exploitation trade-off by independently controverting the convergence rate and the random walk using different parameters – cons EEHO compared to EHO: 1) suitable values for three parameters (rather than two only) have to be found to use EEHO. Originality/value As the original EHO shows fast unjustified convergence toward the origin specifically, the search method adopted in EEHO is more exploitative than the one used in EHO because of the balanced control of exploration and exploitation based on fixing clan updating operator and separating operator. Further, the star discrepancy measure is adopted to quantify the quality of exploration phase of evolutionary algorithms in general. Operator γ that added EEHO allows the successive local and global searching (exploration and exploitation) and helps escaping from local minima that commonly exist in the search space.

Food acquisition and predation avoidance are key drivers of herbivore behaviour. We investigated the interaction of top-down (predator) and bottom-up (food, fire, thermal) effects by measuring the relationship between wolf (Canis lupus L., 1758) predation risk perceived by elk (Cervus elaphus L., 1758) and elk landscape use. We conducted fecal pellet and wolf scat surveys in three valleys with three wolf population levels (Saint Mary: low; Waterton: moderate; North Fork: high). In the North Fork, 90% of quaking aspen (Populus tremuloides Michx.) stands burned recently; the other valleys had no fire. We created predictive models of elk pellet density that incorporated bottom-up and top-down variables. All valleys had a high elk pellet density (≥10 per 100 m2). Wolf scat density was similar where there was no fire, but one order of magnitude greater in burned areas. Elk pellet density was lower in the North Fork, a predation-related response. In all valleys, site-specific elk density declined as impediments to detecting or escaping wolves increased, and elk avoided aspen, except for North Fork unburned areas. Models that best predicted elk density contained bottom-up and top-down effects. At local scales, high predation risk negatively influence elk occurrence, suggesting that even with minimal wolf exposure elk avoid risky sites.

High flow arteriovenous malformations (AVMs) are infiltrative, invading tissue planes and structures and may be life threatening when they bleed.1 They have a feeding artery and an anomalous capillary bed shunting blood from the arterial system to the venous system.1 The present trend of management of small AVMs is surgical excision with a high success rate. The problematic cases are diffuse AVMs infiltrating structures that render them impossible to totally extirpate surgically without causing much blood loss and tissue damage. The mainstay of management is embolization, surgical resection and reconstruction.2 Ligation or proximal embolization (alone) of feeding vessels should never be done because such maneuvers result in rapid recruitment of new vessels from adjacent arteries to supply the AVM nidus.2 Incomplete surgical excision definitely leads to recurrences, making this type of AVM very difficult to manage. What is the point of this paper? The complete destruction of the “nidus” of the AVM, from the artery to the capillary to the venous component, is the only potential cure.3 Well and good if there could be a way of doing this by sclerosing the entire vascular malformation. But since sclerosis only works well in low flow vascular malformations and tumors like hemangiomas, and poorly or not at all in high flow lesions,4 we have to convert this high flow AVM into a “no-flow” or “low flow” AVM by ligating the feeding and draining vessel and injecting the sclerosant intra-arterially thereafter at a dose sufficient enough to blanch out the AVM even up to its peripheral branches. This paper aims to demonstrate how we do this. Definition of Terms Vascular malformation: They are a result of abnormal development of vascular elements during embryonic or fetal stages of life.2 They originate from mesenchymal cells at an early stage of embryogenesis.3 and most are present at birth but there are several case reports of these lesions presenting after trauma in adults.1 Some AVMs appear as part of a familial genetic disorder called angiomatous syndrome i.e. Rendu-Osler-Weber Syndrome presenting with telangiectasia of the skin and mucous membranes.3 Some propose that a defect in vascular stabilization like TGF-beta signaling could be a cause of AVM development.5,6 Still, progesterone receptors have been isolated in AVMs explaining their expansion during puberty.7 Hemangioma: These are vascular tumors that exhibit endothelial proliferation.2 A hemangioma of infancy usually undergoes 3 stages: a proliferative phase of rapid growth up to 10 to 12 months of age; an involuting phase where growth slows down and signs of regression appear usually at 1 to 7 years; and an involuted phase.2 Sclerosants: Agents used in sclerotherapy that induce a toxic effect on the vascular endothelium and results in fibrosis. There are 3 types: Detergents that disrupt cell membranes by protein theft desaturation ie: ethanolamine oleate, sodium morrhuate, polidocanol, sodium tetradecyl sulfate; Osmotic agents ie: sclerodex; and Chemical irritants that damage cell walls by direct contact ie: chromatin glycerine, polyiodinated iodine.8 Sclerodex: an osmotic sclerosant that is a combination of dextrose monohydrate 250mg/ml and sodium chloride 100mg/ml. It shifts water balance through cellular gradient (osmotic) dehydration that leads to endothelial destruction. Since component materials are naturally occurring bodily, it has no molecular toxicity in calibrated dosages. If extravasated, it could cause tissue necrosis.8 It is manufactured by OMEGA Laboratories, Ltd. Montreal, QC, Canada. REVIEW OF PRESENT PRACTICE AND LITERATURE The first task of the physician is to establish a diagnosis, whether the lesion is a vascular malformation or a vascular tumor. Taking the history of the patient could point to a diagnosis as vascular tumors like hemangiomas usually proliferate and involute from the time of infancy to about the age of 10. Vascular malformations grow as the child grows and do not involute. Vascular malformations have an arterial supply and a venous drainage and are classified into high or low flow. Capillary, venous and lymphatic types are low flow while arteriovenous malformations are usually high flow. A high flow AVM has an arterial blood supply and a venous drainage. In rare instances, a vascular malformation could co-exist with a hemangioma forming a mass effect.2 On physical examination, a bruit and a strong pulsation (thrill) is appreciated. The head and neck is the most common location of AVMs at 70%. When fully developed, they are deeper in color with increasing erythema, local warmth, palpable mass and a bruit.9 These malformations are composed of vascular channels lined by flat mature epithelium and are not hypercellular and not proliferative.10 Schobinger proposed a staging system for Head and Neck AVM. Stage 1 are AVMs that are quiescent and remain stable for long periods of time. Stage 2 is a time for expansion followed by pain and bleeding. Stage 3 is heralded by destruction of adjacent tissues and ulceration. Stage 4 is presented by decompensation where symptoms of cardiac failure are present.11 Ultrasound with color Doppler imaging, Magnetic Resonance Imaging and Phlebography (arteriography/venography) contribute to diagnosis, classification and management.3 In our setting were we do not have the facilities, we use CT- angiography. These imaging modalities should be used to evaluate the characteristics of the lesion, such as size, flow velocity, flow direction, relation to surrounding structures and lesion content.3 Ultrasound demonstrates flow rates, contrast-enhanced magnetic resonance imaging (MRI) shows presence or absence of a mass, and CT angiography reveals the arborization (the blood supply and drainage) of the vascular anomaly. Vascular tumors like hemangioma, if located in non-strategic areas where function is not impaired can be observed over its developmental phases until involution at about 10 years old.2 For hemangiomas that impair function or are possibly life threatening because of potential hemorrhage, these tumors are treated with the following modalities: 1. Intralesional corticosteroids ie: triamcinolone; 2.Systemic corticosteroids in a tapered dose like prednisolone and some second choice pharmacotherapeutics like interferon, vincristine; and 3. Propranolol. Surgery is indicated in ulcerating, bleeding, and life threatening lesions like airway obstruction.2 Over 90% demonstrate dramatic reduction in size of hemangioma in one to two weeks from the above medical therapeutic modalities. Propranolol has been successfully used as hemangioma treatment since 2008 and is believed to have an antiproliferative effect on the vascular endothelium. The mechanism of action may involve the regulation of growth factors.1 Low flow vascular malformations are treated with sclerotherapy or surgical excision for accessible tumors. These malformations do not regress like hemangioma but grow in time. High flow AVMs are treated with surgical excision if they have limited extent and are surgically accessible. Embolization before surgery decreases bleeding and is the standard. Embolization followed by repeated sclerotherapy is recommended for surgically inaccessible areas.4 There is a 64 to 96% response rate , defined as improvement in symptoms or a reduction in the lesion size after ethanol sclerotherapy of venous low flow malformations.3 Partial surgical excision leads to only temporary improvement followed by re-expansion of tumor overtime.9 Sometimes, complete resection is not possible in diffuse or infiltrating AVMs and surgery can result in severe disfigurement and impairment of function of involved structures.9 METHODS After establishing the diagnosis of a high flow AVM with identification of an arterial feeding vessel and a venous draining vessel, surgery is commenced away from the malformation to expose the arterial and venous supplies. The procedure is done under general anesthesia because sclerotherapy of large malformations and vascular tumors is very painful. We do this because we have no interventional radiology services in our hospital. We ligate the feeding artery and if possible, the draining vein to convert the AVM into a “ low or no flow” and to allow ample contact time between the sclerosant (sclerogen) and the vessel endothelium. The sclerosant is then injected intravascularly distal to the ligation until all visible malformation blanches out. Aspirating the blood content of the ligated (arterial supply and venous drainage) malformation before introduction of the sclerosant will further potentiate the action of the sclerosant. While injecting the sclerosant slowly, the patient’s vital signs are monitored. A drop in the pulse rate is a signal to stop or slow down the injection of the sclerosant because it may be a sign that some sclerosant is escaping the venous drainage and reaching the general circulation in a concentration picked up by the sensors of the vascular system. Injections resume in a slower manner as the vital signs revert to normal. Our sclerosing agent sclerogen is an osmotic agent composed of sodium chloride and dextrose which are naturally present in our body so they are not toxic in manageable concentrations. Other sclerosants can be nephrotoxic so we must be very careful in injecting not to overload the vascular system. The objective is to push the sclerosant to all branches of the malformation to eliminate all possible nidus. The end point of injection is when all cutaneous or mucosal components of the malformation blanch out. Injection of the sclerosant intravascularly is done under direct visualization to prevent extravascular introduction. A review of literature has this to say about sclerotherapy. Extravascular injection of the sclerosant causes tissue necrosis.4 Ethanol injection to high flow fistulous lesions is contraindicated because of high risk of” early wash “ into the systemic circulation.4 Sclerosants could cause hemolysis, denaturation of blood proteins, thrombus formation and nephrotoxicity.3,4,12 Ethanolamine oleate, in comparison to ethanol, has less effect in the deep vascular layer and no penetrative effect. It is not associated with neuronal side effects despite of the proximity of the nervous system to the vascular system.3 CASES Our first case was a 62-year-old woman with a pinna and periauricular vascular malformation, noted since 5 years prior to consultation. The inferior concha was bulging and pulsating with a strong bruit. The periauricular area was elevated with microvascular malformations in reddish discoloration. The left posterior auricular artery was identified as the feeding artery and the diagnosis was a high flow AVM. She consulted a hospital in Sacramento, California where she worked and was advised to have a resection of her left ear. She got frightened and decided to come home to the Philippines for a second opinion. I suggested our procedure which she gladly accepted but warned her of possible pinna necrosis. At least, she said, it is just a possibility and not an outright pinna loss. I dissected 1 cm below the malformation avoiding any of its extensions below the pinna and mandibular angle and moved towards the external carotid. I immediately located the pulsating, abnormally dilated posterior auricular artery feeding vessel and ligated it. Further dissection deep towards the styloid process revealed the venous drainage that penetrated the mastoid bone toward the direction of the sigmoid sinus. I too did the venous drainage ligation. I injected sclerogen distal to the posterior artery ligation after aspiration of 8 cc of AVM blood until all the malformation main mass and the peripheral branches blanched out. Total volume of sclerosant was 10 cc. There was no change in the vital signs as I slowly introduced the sclerosant. I closed the surgical defect and observed the patient for three days in the hospital. There was post-operative pain and swelling in the sclerosed malformation, relieved by ice packs and celecoxib 200 mg every 12 hours. After three days, the swelling started to subside and the pain lessened so the patient was sent home. She followed up in a week and the malformation had shrunk. Sutures were removed. Two weeks post-operatively, the malformation was just a trace skin discoloration with no tissue necrosis, no more bulge and pulsations and no pain. She asked permission to go back to work the following week in California. Our second case was a 13-year-old girl with a right tonsillar and hypopharyngeal vascular malformation. She had recurrent bleeding episodes necessitating blood transfusions in their province. I suggested our procedure which the parents and the patient consented to. We did surgery, ligating the right external carotid artery and external jugular vein and introduced the sclerosant (sclerogen) slowly until the tonsillar and hypopharyngeal malformation blanched out. There was no abnormal fluctuation of the patient’s vital signs. After closing the surgical access wound, I did tonsillectomy of the right since the bulging tonsillar malformation was obstructing the airway. There was very minimal bleeding and I was able to cauterize the remaining sclerosed malformation not included in the tonsil with ease. Two weeks post-surgery, she followed up with healed tonsillectomy wound and a disappearing malformation. She however had gastritis because of her co-amoxiclav antibiotic and her inability to eat well because of pain in swallowing. She eventually recovered from her gastric problems. At one-month follow-up, there was no trace of the malformation on visual examination. Looking back, doing tonsillectomy in an AVM would have been very bloody without sclerotherapy. Our third case was a nasopharyngeal AV malformation in a 35-year-old woman. She had episodes of severe bleeding requiring emergency tracheostomy, oral packing and blood transfusions. CT-angiography revealed two feeding vessels, one from the left external carotid artery and a minor one from the internal carotid artery. We decided to sclerose the left external carotid artery and see what happened to the internal carotid artery branch that could not be accessed. Since CT angiography did not identify the venous drainage, we introduced the sclerosant (sclerogen) very slowly, stopping when the pulse rate started to drop below 60 beats per minute and resuming slow injection when the pulse rate was normal. Oxygen saturation was noted to be stable at 98 to 100 %. We stopped when the AVM blanched out, injecting 15 ml of sclerosant. In two weeks time the AVM shrank except for a 1 x 1 cm bulge at the left posterior nasopharynx that was supplied by the internal carotid artery branch that could not be sclerosed at the time of surgery. The patient was decannulated from tracheostomy and was able to resume normal diet and activity. She is on regular follow-up and is being maintained with propranolol 40 mg once a day hoping that it may work as it does in hemangioma. 1 year post operatively, the bulge has not grown nor disappeared. Looking back, had we done surgery as suggested by colleagues, we could have encountered massive bleeding, inability to take the AVM all out, and eventual recurrence. Pondering upon the case of a second arterial blood supply of the AVM, the malformation could have recruited this second blood supply. The forward force of introduction of the sclerosant was not able to overcome the arterial pressure of the internal carotid artery feeding branch so the sclerotherapy effect stopped where the flow forces where at equilibrium. Note that in this case, we did not ligate the specific venous drainage as the CT-angiography did not identify it. DISCUSSION Managing AVMs that are diffuse and infiltrative can be very difficult. Surgical extirpation of all the nidus may not be possible and will surely lead to recurrences. Besides, malformations located in functionally strategic areas may present with structural deformities and functional disturbances when they are damaged by surgery. Small AVMs can be resected with high rates of success and no recurrences. In one series, all 16 patients with surgically accessible, localized, non-infiltrating AVMs who underwent angio-embolization with subsequent surgical excision demonstrated no evidence of recurrence on angiography during follow-up averaging 3 months.9 In low flow venous malformations, sclerotherapy administered by trans-arterial, trans-venous or direct puncture injection without embolization or feeding vessel ligation has a 64 to 96% response rate, defined as improvement in symptoms or reduction of the lesion and not necessarily cure.3 The cure rate for small malformations was 69% with excision only and 62% for extensive lesions with combined embolization and resection. At 6 years average follow-up, cure rate was 75% for stage 1; 67% for stage 2; and 48% for stage 3 malformations. The outcomes were not significantly affected by age at treatment, Schobinger stage, or treatment method.11 Embolo-sclerotherapy is a new therapeutic modality for surgically inaccessible lesions like diffuse and infiltrating AVMs.4 This procedure is done repeatedly since the embolus recanalizes and the sclerosant is injected distal to the embolus. This method is reserved as an adjunct to subsequent surgical resection.4 Our immediate results for ligation of feeding artery or draining vein before sclerotherapy were dramatic without functional or anatomic compromise. With sclerogen, whose components exist in the body naturally, we found no significant complications in our 3 cases. This technique could be ideal for diffuse and infiltrating AVMs. It is more effective if the feeding artery and the draining vein are identified and ligated so that the sclerosant can be pushed to all branches of the AVM. With a “no-flow or a slow-flow” AVM, we are able to prolong the contact of the sclerosant with the vascular endothelium thereby increasing the success rate of totally eliminating the nidus of the vascular malformation. More studies and experience are needed to prove the durability of this technique. Are we able to eliminate all the nidus of the AVM if we are able to ligate all feeding arteries and draining veins before sclerotherapy? Our center is not so equipped so more technically advanced institutions dealing with vascular tumors and malformations can validate the efficacy of this technique. After all, it might not be bane to ligate AVM feeding and draining vessels if we are able to destroy the entire nidus of the AVM by whatever means like sclerotherapy in this case. Acknowledgements I wish to acknowledge my anesthesiologist Dr. Julius Apostol who encouraged me to try new things and promised to research on management of possible egress of sclerosants in the systemic circulation as he puts my patients to sleep. My residents at the Department of Otolaryngology-Head and Neck Surgery, Baguio General Hospital did the pre- and post-operative patient care while I was away, and the photography: Dr. Carlo Pagalilauan, Chief Resident and photographer, Dr. Sherwin Valdez, Dr. Beverly Carbonel, Dr. Jeff Peckley and Dr. Wingleaf Yu who are my assistants. Thank you.

AbstractArtificial neural network (ANN) has been commonly used to deal with many problems. However, since this algorithm applies backpropagation algorithms based on gradient descent (GD) technique to look for the best solution, the network may face major risks of being entrapped in local minima. To overcome those drawbacks of ANN, in this work, we propose a novel ANN working parallel with metaheuristic algorithms (MAs) to train the network. The core idea is that first, (1) GD is applied to increase the convergence speed. (2) If the network is stuck in local minima, the capacity of the global search technique of MAs is employed. (3) After escaping from local minima, the GD technique is applied again. This process is applied until the target is achieved. Additionally, to increase the efficiency of the global search capacity, a hybrid of particle swarm optimization and genetic algorithm (PSOGA) is employed. The effectiveness of ANNPSOGA is assessed using both numerical models and measurement. The results demonstrate that ANNPSOGA provides higher accuracy than traditional ANN, PSO, and other hybrid ANNs (even a higher level of noise is employed) and also considerably decreases calculational cost compared with PSO.

AbstractOil formation volume factor (FVF) is considered as relative change in oil volume between reservoir condition and standard surface condition. FVF, always greater than one, is dominated by reservoir temperature, amount of dissolved gas in oil, and specific gravity of oil and dissolved gas. In addition to limitations on reliable sampling, experimental determination of FVF is associated with high costs and time-consumption. Therefore, this study proposes a novel approach based on hybrid genetic algorithm-pattern search (GA-PS) optimized neural network (NN) for fast, accurate, and cheap determination of oil FVF from available measured pressure-volume-temperature (PVT) data. Contrasting to traditional neural network which is in danger of sticking in local minima, GA-PS optimized NN is in charge of escaping from local minima and converging to global minimum. A group of 342 data points were used for model construction and a group of 219 data points were employed for model assessment. Results indicated superiority of GA-PS optimized NN to traditional NN. Oil FVF values, determined by GA-PS optimized NN were in good agreement with reality.

Several well known clustering algorithms have their own online counterparts, in order to deal effectively with the big data issue, as well as with the case where the data become available in a streaming fashion. However, very few of them follow the stochastic gradient descent philosophy, despite the fact that the latter enjoys certain practical advantages (such as the possibility of (a) running faster than their batch processing counterparts and (b) escaping from local minima of the associated cost function), while, in addition, strong theoretical convergence results have been established for it. In this paper a novel stochastic gradient descent possibilistic clustering algorithm, called O- PCM 2 is introduced. The algorithm is presented in detail and it is rigorously proved that the gradient of the associated cost function tends to zero in the L 2 sense, based on general convergence results established for the family of the stochastic gradient descent algorithms. Furthermore, an additional discussion is provided on the nature of the points where the algorithm may converge. Finally, the performance of the proposed algorithm is tested against other related algorithms, on the basis of both synthetic and real data sets.

To listen to them talk, you'd think most Americans hate television. Everyday discourse about television abounds with condemnation of television content. Television is a wasteland, a stream of idiotic material insulting to the intelligence of the viewer. When people deem a particular program worth watching, they often articulate it in contradistinction to the vast majority of awful stuff out there. This almost universal discourse of condemnation does not mean Americans do not watch television, of course. They do, and they watch a great deal of it. Thus we have a conundrum. If it is so awful, why do people watch television? When Americans construct stories about themselves, they construct themselves as choice--making individuals (Polanyi). Sane, mature Americans are expected to be able to make intelligent choices and to live with the consequences of their choices. How, then, can Americans articulate themselves as television viewers, as individuals who choose to view what is clearly awful stuff? In this paper, I want to discuss 'veging out' as an American category of media viewing that resolves this conundrum. In framing their discourse about watching television in terms of 'veging out,' Americans are able to construct themselves as sensible, choice--making persons, and yet explain why they watch large amounts of television. I want to use this example to explore ways that media scholars might supplement explorations of the self as mediated by texts with attention to the ways the viewing self is articulated in everyday discourses about television by viewers. An American Folk Category of Pleasure I said I'm sorry this is late. I just couldn't work on it over the weekend. I just veged out in front of the TV the whole weekend. I realise that's not much of an excuse…but…I had my Arabic test Thursday and I was too burned out afterward to do anything. I had to let my brain recharge. [text one] Let's just veg out tonight. We both had a big lunch, let's just make some popcorn and watch whatever stupid stuff is on TV. Unless you want to get a video. [text two] God, we didn't do anything this weekend. We just sat in front of the TV. (laughs) It was a total veg out weekend, we ordered out every night. John was on the rig for two weeks and then he's had to work late every night since he's been back, and I've had this activity and that activity with the kids, and girl scouts and soccer... We really needed the break. [text three] In the interest of brevity, I offer only three texts here.1 Anyone who has listened to Americans talk about television can probably multiply these examples many times; most Americans of my generation or later have almost certainly been producers of such discourse at one time or another. Each of these examples is drawn from a different context: a student's explanation for handing in a late paper [text one], a wife's suggestion for evening plans [text two], a friend sharing information about her family [text three]. And each is part of the language of experience – the language people use to describe emotions, sensations, and thoughts and, in so doing, articulate a self. 'Veging out' -- the 'veg--' prefix is borrowed from the word 'vegetable' and pronounced with a soft g -- is a nice example of a local taxonomic category of pleasure and the way it is embedded in more complex discursive formations, which it both replicates and refracts. In American society, where sitting in front of the television when there are other things to do is condemned as a waste of time that makes one a 'couch potato,' 'veging out' allows actors to reconstitute 'being a vegetable' as an empowering choice, an intentional and temporary vegetative state one escapes into as a means to relax, reduce stress and 'get away' from one's troubles. Veging out involves escape but specifies that one is escaping to nowhere, that an avoidance of critical mental activity is precisely what is sought. The claim to be veging out thus accepts the general American public discourse of television as a wasteland – the 'waste' in particular involving waste of time -- and simultaneously challenges it by claiming, in essence, that one has a right to do nothing if one has been working 'too hard'. There is nothing fanciful or even insightful in this analysis; discourses in which Americans talk about their television viewing activity tend to be both straightforward and redundant. Americans who say they spent an evening veging out are likely to follow the statement with an explanation of why they are entitled to veg out -- a litany of stresses or labours -- and sometimes also assertions to confirm that the world they escaped to was indeed a place that involved minimal mental activity. For example, the student in Text One quoted above followed it up with the comment, 'There was absolutely nothing on worth watching'. The woman who produced Text Three commented a few lines later, 'It was practically all commercials, nothing could hold my interest because it was always being interrupted. I hardly ever watch TV, I hadn't realised how many commercials there are'. This latter comment also positions the activity as a rare one for this person, emphasising the strategic nature of veging out as a life choice and hence acceptable within American understandings of choice.2 People's own modes of articulation may thus even deny their motivations involve pleasure.3 Choosing to enter the wasteland of television certainly can be, and often is, constructed as a bad choice. As Beeman demonstrates in his analysis of the language of choice in American advertising, making a choice is often constituted as not enough -- one must make the 'right' choice. Discourse about 'veging out' partly forecloses the possibility of the instance described being a bad choice by embedding the choice in the matrix of suffering. Yet as Carbaugh discovers in his sociolinguistic appraisal of TV talk shows, doing something 'wrong' can nonetheless be valorised in America by its formulation as a deliberate exercise of one's right to choose. The moral wrongness of the particular choice is redeemed by the articulation of a self exercising its right to make its own choices, and taking responsibility for those choices. The power of 'veging out' as a representation of social action thus lies in its ability to simultaneously embrace the widespread discourse that 'television is a wasteland' while at the same time subsuming it under the important American discourse of choice. In so doing, it allows Americans to construct themselves as hard--working individuals who choose to waste time as a strategy for resolving the stresses and discomforts of hard work. One articulates a viewing self, that is, which is consonant with the fundamental values of American culture. The Viewing Self The 'viewing self' is that self, or that aspect of the self, constructed through experiences of viewing events and activities in which the person is not a participant. In the contemporary world, such viewing has increased as an activity, accommodated and mediated by film, television, video and other technologies. These technologies offer, among other things, the opportunity for virtual experiences, events and activities that we do not experience with our bodies but which nonetheless offer us comparable fodder for our cognitive processes (Drummond). Studies of the self as viewer have long been dominated in media studies by attention to these virtual experiences as internal. From the early argument that the self is 'interpellated' by the culture industry (Adorno), to the argument that the self is socially and politically positioned in dominated, negotiating or resistant ways (Hall), to the idea of the self as simultaneously occupying multiple (and shifting) spectator positions (Modleski, Williams, Clover, Caton), emphasis has long been on how the viewer experiences structured sets of symbols, appropriates them at various levels of cohesion, cognitively and affectively orders them with regard to pre--existing understandings of and feelings about the world, and uses them in the ongoing construction of the self. I am suggesting here the utility of turning our attention from internal to external articulations of self as viewer. I want to argue that in addition to engaging with the content of the viewing experience, people usually engage with the meaning of the viewing experience as an activity. The viewing experience is never just about engagement with content about what one watches. It is also about the activities of 'watching TV,' 'renting a video,' and 'going to the movies.' Each of these is an experience that must be internally evaluated with regard to one's pre--existing sense of self, and which may have to be verbally articulated in interaction with others. In the latter case, it provides yet more fodder for the construction of the self, as we see versions of ourselves mirrored in the responses of the other to our own self--performance. Given the plethora of media, genres, places and events in which visual media are watched, speaking with others about one's television viewing maps one onto a complex terrain of distinctions about one's taste. One's 'taste' is never innocent, because it ties in to a complex social code that relates it to class, gender, ethnicity, education, and other social categories (Bourdieu). To represent ourselves to others as viewers of any particular kind of media is to position ourselves as particular kinds of persons in relation to others. One can use this code to articulate oneself as a particular kind of person vis--à--vis those with whom one is interacting: an equal who shares common tastes, a superior who enjoys more refined discernment, a populist who revels in his or her common tastes. To speak of our viewing allows us to generate social contact on grounds of shared experience. It allows us to confirm our tastes with regard to the social others who serve as mirrors to our selves. Of course, persons are never omnicompetent in their self--presentations, and efforts to present the self in particular ways can backfire, so that instead of appearing as a woman of discernment one appears pompous; and instead of appearing as a common Joe, one comes across as vulgar. Talking about viewing, in other words, always involves risk. In examining how people manage this risk in their social interactions, as through framing their experience as 'veging out,' we can learn much about how people construct themselves as viewers. Conclusion 'Veging out' is not the only verbal strategy by means of which Americans solve the conundrum of the viewing self. Nor is there anything unique in this American conundrum. Ethnographic accounts clearly demonstrate that many societies offer public condemnatory discourses about television that are at odds with actual viewing practices. The content of television in Belize is 'destroying a whole generation' (Wilk), in Egypt it's a flood of 'moral pollution' (Armbrust), in the Netherlands it's 'an embarrassment' (Alasuutaari). People's ways of speaking about themselves as viewers are clearly often a result of an ambivalence born of their pleasure, on the one hand, and their understanding that one should not be getting pleasure from such stuff, on the other. The result is often discourse that expresses guilt, or embarrassment, as summed up by Alasuutari's informant who said 'I'm ashamed to admit it, but I watch Dallas.' Alasuutari's reliance on interviewing, though, captures the conundrum but not the cultural solutions. An interview with a sociologist is a very different kind of speech act from the quotidian contexts in which people construct themselves as television viewers in interaction with friends, family, the person sitting next to you at the bar, and so forth (Briggs). My objective in this brief exercise is to draw our attention away from interviewing toward ethnography, and away from attention to internal subjectivities to the interactive contexts in which the self is constructed in everyday life. Notes 1 These three examples were all collected among American expatriates while I was teaching at the American University in Cairo. 2 Individual performances of this discourse are always strategic, of course; their articulation shaped by the speakers understanding of the speech event in which they take place. 3 The American discomfort with spending one's leisure pleasurably has been long chronicled. As early as the 1920s the Lynds found the people of Middletown uncomfortable with talking about reading for pleasure rather than instruction and profit. People did not want to articulate themselves as persons who wasted time (Lynd and Lynd 1929: 225) References Adorno, Theodor. 'The Culture Industry Reconsidered.' The Adorno Reader. Ed. Brian O'Connor. Oxford: Blackwell, 2000. 230-38. Alasuutari, Pertti. ''I'm Ashamed to Admit it, but I have Watched Dallas:' The Moral Hierarchy of Television Programmes.' Media, Culture and Society 14 (1992): 561-582. Armbrust, Walter. Mass Culture and Modernisation in Egypt. Cambridge: University Press, 1996. Beeman, William O. 'Freedom to Choose: Symbolic Values in American Advertising.' The Symbolisation of America. Ed. Herve Varenne. Omaha: University of Nebraska Press, 1986 Bourdieu, Pierre. Distinction: A Social Critique of the Judgment of Taste. Trans. Richard Nice. Harvard: Harvard University Press, 1984. Briggs, Charles. Learning How to Ask: A Sociolinguistic Appraisal of the Role of the Interview in Social Science Research. Cambridge: Cambridge University Press, 1986. Carbaugh, Donal. Talking American: Cultural Discourses on Donahue. Norwood, NJ: Ablex, 1989. Caton, Steven C. Lawrence of Arabia: a Film's Anthropology. Berkeley: University of California Press, 1999. Clover, Carol. Men, Women, and Chain Saws: Gender in the Modern Horror Film. Princeton, NJ: Princeton University Press, 1992. Drummond, Lee. American Dreamtime: A Cultural Analysis of Popular Movies and Their Implications for a Science of Humanity. Lanham, MD: Littlefield Adams, 1995. Hall, Stuart. 'Culture, the Media and the 'Ideological Effect.' ' Mass Communication and Society. Ed. James Curran, Michael Gurevitch and Janet Woolacott. London: Edward Arnold, 1977. - - - . 'The Rediscovery of 'Ideology:' The Return of the Repressed in Media Studies. Culture, Society and the Media. Ed. Michael Gurevitch, T. Bennett, James Curran and Janet Woolacott. London: Methuen, 1982. Modleski, Tania. The Women Who Knew Too Much: Hitchcock and Feminist Analysis. New York: Routledge, 1988 Polanyi, Livia. Telling the American Story. Cambridge, Mass: MIT Press, 1989 Wilk, Richard. ''It's Destroying a Whole Generation:' Television and Moral Discourse in Belize.' Visual Anthropology 5 (1995): 229-44. Williams, Linda. Hard Core: Power, Pleasure and the Frenzy of the Visible. Berkeley, CA: University of California Press, 1989. Citation reference for this article Substitute your date of access for Dn Month Year etc... MLA Style Peterson, Mark Allen. "Choosing the Wasteland" M/C: A Journal of Media and Culture 5.5 (2002). [your date of access] < http://www.media-culture.org.au/mc/0210/Peterson.html &gt. Chicago Style Peterson, Mark Allen, "Choosing the Wasteland" M/C: A Journal of Media and Culture 5, no. 5 (2002), < http://www.media-culture.org.au/mc/0210/Peterson.html &gt ([your date of access]). APA Style Peterson, Mark Allen. (2002) Choosing the Wasteland. M/C: A Journal of Media and Culture 5(5). < http://www.media-culture.org.au/mc/0210/Peterson.html &gt ([your date of access]).