Improving reinforcement learning by using sequence trees

This paper proposes a novel approach to discover options in the form of stochastic conditionally terminating sequences; it shows how such sequences can be integrated into the reinforcement learning framework to improve the learning performance. The method utilizes stored histories of possible optimal policies and constructs a specialized tree structure during the learning process. The constructed tree facilitates the process of identifying frequently used action sequences together with states that are visited during the execution of such sequences. The tree is constantly updated and used to implicitly run corresponding options. The effectiveness of the method is demonstrated empirically by conducting extensive experiments on various domains with different properties.     

HEAT AND MASS TRANSFER IN IMPINGEMENT DRYING

ABSTRACT

In industrial drying applications, efficient transfer of heat and mass between a drying medium and the material being dried is very critical for the overall economics of the operation. Impinging jets are therefore widely used for their enhanced tmnsport characteristics, especially for drying of continuous sheets of materials such as paper and textiles. In such applications, a thin sheet of material, as wide as 6m in cross machine direction, speeds at velocities as high as 90 km/hr under high velocity jets emerging from a confining surface parallel to the material surface. Many variables and effects need to k considered for proper design of such impinging jet systems: nozzle geometry and size, nozzle configuration, location of exhaust pons, nozzle-to-surface separation, jet-to-jet separation, cross flow, jet exit velocity and surface motion. For permeable materials, additional enhancement of heat and mass transfer that occur when some of the impinging gas is removed through the material makes this option an atmctive one.

Here, we review the above effects and offer predictive correlations from literature which may be used in the design of high velocity impinging jet systems.     

Lumped-parameter models for low-temperature geothermal fields and their application

The behavior of low-temperature geothermal reservoirs under exploitation is simulated using analytical lumped-parameter models. These models consider the effects of fluid production and reinjection, as well as natural recharge, on the pressures (or water levels) of low-temperature, liquid-dominated geothermal systems. The computed responses for constant production/injection flow rates are given in the form of analytical expressions. Variable flow rate cases are modeled, based on the Duhamel's principle. Reservoir parameters are obtained by applying a weighted nonlinear least-squares estimation technique in which measured field data are history matched to the corresponding model response. By using history-matched models, the future performance of the reservoir can be predicted for different production/injection scenarios in order to optimize the management of a given geothermal system.We demonstrate the applicability of the models by simulating measured data from the Laugarnes geothermal field in Iceland, and the Balcova–Narlidere field in Turkey.     

Diagnosis of heart disease using artificial immune recognition system and fuzzy weighted pre-processing

This paper presents a novel method for diagnosis of heart disease. The proposed method is based on a hybrid method that uses fuzzy weighted pre-processing and artificial immune recognition system (AIRS). Artificial immune recognition system has showed an effective performance on several problems such as machine learning benchmark problems and medical classification problems like breast cancer, diabetes, liver disorders classification. The robustness of the proposed method is examined using classification accuracy, k-fold cross-validation method and confusion matrix. The obtained classification accuracy is 96.30% and it is very promising compared to the previously reported classification techniques.     

Physical properties of CdS:Ga thin films synthesized by spray pyrolysis technique

This paper reports the investigation of physical properties of CdS:Ga thin films grown for the first time by a simple spray pyrolysis method as a function of Ga-doping level from 0 to 8 at.%. X-ray diffraction (XRD), scanning electron microscopy (SEM), energy-dispersive photoelectron spectroscopy, transmittance, photoluminescence, Hall effect and resistivity measurements are utilized to search for the structural, morphological, chemical, optical and electrical properties of as-prepared samples. XRD data confirm the presence of hexagonal structure with a strong (101) preferred orientation. SEM results show that the surface morphology varies significantly via Ga-doping, particularly 6 at.% doping level. Optical transparency is improved by the lower Ga-doping (2 and 4 at.%) whereas higher doping concentration (6 and 8 at.%) causes a poor transmission in the visible region. With respect to CdS (2.42 eV), the calculated band gap values at first enhances for 2 at.% Ga-doping and reaches to 2.43 eV. But, further increase in Ga-doping amount leads to a drop in the band gap value (2.39 eV) for 8 at.% Ga-doping. Electrical analyses display that 2 at.% Ga-doped CdS thin films exhibit a maximum carrier density and a minimum resistivity that are related to the substitutional incorporation of Ga³⁺ ions at Cd²⁺ ions. However, higher doping of Ga atoms into CdS gives rise to a gradual diminish in the carrier concentration and a rise in the resistivity. Based on all the data, it should be concluded that 2 at.% Ga-doped CdS thin films exhibit the best optical and electrical properties that can be used in the optoelectronic applications.     

An iterative genetic algorithm for the assembly line worker assignment and balancing problem of type-II

In this study, we consider the assembly line worker assignment and balancing problem of type-II (ALWABP-2). ALWABP-2 arises when task times differ depending on operator skills and concerns with the assignment of tasks and operators to stations in order to minimize the cycle time. We developed an iterative genetic algorithm (IGA) to solve this problem. In the IGA, three search approaches are adopted in order to obtain search diversity and efficiency: modified bisection search, genetic algorithm and iterated local search. When designing the IGA, all the parameters such as construction heuristics, genetic operators and local search operators are adapted specifically to the ALWABP-2. The performance of the proposed IGA is compared with heuristic and metaheuristic approaches on benchmark problem instances. Experimental results show that the proposed IGA is very effective and robust for a large set of benchmark problems.     

A new hybrid recommendation algorithm with privacy

Providing accurate and dependable recommendations efficiently while preserving privacy is essential for e‐commerce sites to recruit new customers and keep the existing ones. Such sites might be able to increase their sales and profits while customers can obtain precise and trustworthy predictions if they use appropriate collaborative filtering (CF) algorithms without deeply jeopardizing users' privacy. We propose a new recommendation algorithm, which is a hybrid‐memory and model‐based algorithm to generate truthful referrals efficiently. Moreover, we use randomization techniques to preserve users' privacy while still offering CF services with decent accuracy. We perform real data‐based trials and analyse our proposed schemes in terms of privacy, accuracy, and performance.     

Automatic determination of traffic accidents based on KMC-based attribute weighting

In this study, the traffic accidents recognizing risk factors related to the environmental (climatological) conditions that are associated with motor vehicles accidents on the Konya-Afyonkarahisar highway with the aid of Geographical Information Systems (GIS) have been determined using the combination of K-means clustering (KMC)-based attribute weighting (KMCAW) and classifier algorithms including artificial neural network (ANN) and adaptive network-based fuzzy inference system (ANFIS). The dynamic segmentation process in ArcGIS9.0 from the traffic accident reports recorded by District Traffic Agency has identified the locations of the motor vehicle accidents. The attributes obtained from this system are day, temperature, humidity, weather conditions, and month of occurred traffic accidents. The traffic accident dataset comprises five attributes (day, temperature, humidity, weather conditions, and month of occurred traffic accidents) and 358 observations including 179 without accident and 179 with accident. The proposed comprises two stages. In the first stage, the all attributes of dataset have been weighted using KMCAW method. The aims of this weighting method are both to increase the classification performance of used classifier algorithm and to transform from linearly non-separable traffic accidents dataset to a linearly separable dataset. In the second stage, after weighting process, ANN and ANFIS classifier algorithms have been separately used to determine the case of traffic accidents as with accident or without accident. In order to evaluate the performance of proposed method, the classification accuracy, sensitivity, specificity and area under the ROC (Receiver Operating Characteristic) curves (AUC) values have been used. While ANN and ANFIS classifiers obtained the overall prediction accuracies of 53.93 and 38.76%, respectively, the combination of KMCAW and ANN and the combination of KMCAW and ANFIS achieved the overall prediction accuracies of 74.15 and 55.06% on the prediction of traffic accidents. The experimental results have demonstrated that the proposed attribute weighting method called KMCAW is a robust and effective data pre-processing method in the prediction of traffic accidents on Konya-Afyonkarahisar highway in Turkey.