Heuristic Training Research Articles

Self-refining games using player analytics

Data-driven simulation demands good training data drawn from a vast space of possible simulations. While fully sampling these large spaces is infeasible, we observe that in practical applications, such as gameplay, users explore only a vanishingly small subset of the dynamical state space. In this paper we present a sampling approach that takes advantage of this observation by concentrating precomputation around the states that users are most likely to encounter. We demonstrate our technique in a prototype self-refining game whose dynamics improve with play, ultimately providing realistically rendered, rich fluid dynamics in real time on a mobile device. Our results show that our analytics-driven training approach yields lower model error and fewer visual artifacts than a heuristic training strategy.

Heuristic ternary error-correcting output codes via weight optimization and layered clustering-based approach.

One important classifier ensemble for multiclass classification problems is error-correcting output codes (ECOCs). It bridges multiclass problems and binary-class classifiers by decomposing multiclass problems to a serial binary-class problems. In this paper, we present a heuristic ternary code, named weight optimization and layered clustering-based ECOC (WOLC-ECOC). It starts with an arbitrary valid ECOC and iterates the following two steps until the training risk converges. The first step, named layered clustering-based ECOC (LC-ECOC), constructs multiple strong classifiers on the most confusing binary-class problem. The second step adds the new classifiers to ECOC by a novel optimized weighted (OW) decoding algorithm, where the optimization problem of the decoding is solved by the cutting plane algorithm. Technically, LC-ECOC makes the heuristic training process not blocked by some difficult binary-class problem. OW decoding guarantees the nonincrease of the training risk for ensuring a small code length. Results on 14 UCI datasets and a music genre classification problem demonstrate the effectiveness of WOLC-ECOC.

Quality improvement of technical education in Saudi Arabia: self-evaluation perspective

Purpose – In Saudi Arabia, technical education is managed by Technical and Vocational Training Corporation. However, there is no independent accrediting body to accredit the technical institutes and assure its quality, which causes the lack of unified quality assurance standards and manageable quality improvement processes. The purpose of this paper is to propose self-evaluation standards to help the technical institutes to evaluate their performance. Design/methodology/approach – The author has used a brainstorming technique of local practitioners in quality assurance. Such technique was performed by adopting DACUM that stands for “Developing A CurriculUM”. The author has used a heuristic educational and training process as the foundation of the standards and processes development, namely; curriculum, environment, training managements, and instructors. A team has been established that has been trained locally and abroad on the quality assurance standards and processes. The team studied several quality frameworks of the different countries to come up with guidelines for quality self-evaluation and standards. Findings – The author has found that such an approach is a very effective tool for improving the institutes' performance and gives them the flexibility to decide about their missions. Most of institutes' staff is reluctant to participate in the self-evaluation process because of the fear it may reveal their weaknesses, but with encouragement and motivation especially from the top management they tend to participate in such a process. Self-evaluation helps the educational institutes to be benchmarked with other international institutions, in which good practices may be adopted by the institutes' managements to achieve their vision. Research limitations/implications – The proposed approach can help the technical education institutes to manage their quality system. In addition, the implementation of such an approach might be the starting point to develop a quality system framework for the technical education in Saudi Arabia. However, further investigation is needed to measure whether applying such standards may help the technical education institutes to meet the required quality standards to attain an accreditation from the international quality agencies. Originality/value – The proposed guidelines for quality standards and processes is a contribution in the accreditation and quality assurance processes for many public and private institutions in Saudi Arabia. In addition, it is an important step to standardize the quality processes.

A Context for Training - Based Approximation in Randomly Correlated Rician Mimo Channels with Rician Disruption

An antenna array on the transmit side provides the system with an extra spatial dimension that can be utilized for coding both in the spatial as well as the temporal domain. The recent development of such space time codes shows that there are ways of exploiting multiple transmit antennas while completely avoiding traditional beam-forming techniques need of accurate channel state information. In this project, we generate a framework for training based channel estimation under different channel and interference statistics. The minimum mean square error (MMSE) estimator for channel matrix estimation in Rician fading multi-antenna systems is analyzed, and exclusively the proposal of mean square error (MSE) minimizing training structures. By considering Kronecker-structured systems with a grouping of noise and interference and random training sequence length, we gather and simplify numerous earlier results in the framework. We simplify the circumstances for attaining the optimal training sequence structure and show when the spatial training power distribution can be explained unambiguously. We also prove that spatial correlation improves the estimation performance and establish how it determines the optimal training sequence length. The analytic results for Kronecker-structured systems are used to derive a heuristic training sequence under general unstructured statistics. The MMSE estimator of the squared Frobenius norm of the channel matrix is also derived and shown to provide far better gain estimates than other approaches. It is shown under which conditions training sequences that minimize the non-convex MSE can be derived explicitly or with low complexity. Numerical examples are used to evaluate the performance of the two estimators for different training classifications and system statistics. We also elucidate how the finest length of the training sequence often can be shorter than the number of transmit antennas. Keyword: mse, mmse estimator, kronecker-structured systems.

Visual tracking via weakly supervised learning from multiple imperfect oracles

Notwithstanding many years of progress, visual tracking is still a difficult but important problem. Since most top-performing tracking methods have their strengths and weaknesses and are suited for handling only a certain type of variation, one of the next challenges is to integrate all these methods and address the problem of long-term persistent tracking in ever-changing environments. Towards this goal, we consider visual tracking in a novel weakly supervised learning scenario where (possibly noisy) labels but no ground truth are provided by multiple imperfect oracles (i.e., different trackers). These trackers naturally have intrinsic diversity due to their different design strategies, and we propose a probabilistic method to simultaneously infer the most likely object position by considering the outputs of all trackers, and estimate the accuracy of each tracker. An online evaluation strategy of trackers and a heuristic training data selection scheme are adopted to make the inference more effective and efficient. Consequently, the proposed method can avoid the pitfalls of purely single tracking methods and get reliably labeled samples to incrementally update each tracker (if it is an appearance-adaptive tracker) to capture the appearance changes. Extensive experiments on challenging video sequences demonstrate the robustness and effectiveness of the proposed method.

The Context of Knowledge and Data Discovery in Highly Dense Data Points Using Heuristic Approach

In data mining framework, for proficient data examination recent researchers utilized branch-and-bound methods such as seriation, clustering, and feature selection. Conventional cluster search was completed with diverse partitioning schemes to optimize the cluster pattern. Considering image data, partitioning approaches seems to be computationally complex due to large data size, and uncertainty of number of clusters. Recent work presented a new version of branch and bound model called model selection problem, handles the clustering issues more efficiently. The existing work deployed spatially coherent sampling for generating cluster parameter candidates. But if the problem-specific bounds and/or added heuristics in the data points of the domain area get surmounted, memory overheads, specific model selection, and uncertain data points cause various clustering abnormalities. To overcome the above mentioned issues, we plan to present an Optimal Model-Selection Clustering for image data point analysis in the context of knowledge and data discovery in highly dense data points with more uncertainty. In this work, we are going to analyze the model selection clustering which is first initiated through the process of heuristic training sequences on image data points and appropriates the problem-specific characteristics. Heuristic training sequences will generate and test a set of models to determine whether the model is matched with the characteristics of the problem or not. Through the process of heuristic training sequences, we efficiently perform the model selection criteria. An experimental evaluation is conducted on the proposed model selection clustering for image data point using heuristic approach (MSCHA) with real and synthetic data sets extracted from research repositories (UCI) and performance of the proposed MSCHA is measured in terms of Data point density, Model-Selection Criteria, Cluster validity.

Methodological concerns in usability evaluation of software prototypes

Karahoca et al. [1] addressed an important topic in their paperconcerning the use of usability evaluation methods to choose anappropriate software prototype for tablet personal computers.They applied a combination of usability evaluation methods toevaluate the usability of two software prototypes with differentgraphical user interfaces (GUIs); iconic and non-iconic. These GUIswere designed to replace the paper-based forms at an emergencydepartment of a Turkish hospital. The whole healthcare staff ofthe department, consisting of 6 physicians and 32 nurses, partici-patedas evaluatorsin the study.The findingsof the comprehensiveevaluation showed that the iconic GUI prototype had better usabil-ity thanthe non-iconicGUI prototype.This studycontributesto thebody of knowledge concerning the usability of GUIs. Such studiesare important from a practical perspective because in the compet-itive market of clinical software they help health care organiza-tions in selecting systems that best suits their users’ needs.Moreover, the results of these studies provide practical input tosoftware developers concerning the design of software that is easyto use and that fits the workflow of healthcare providers.Although Karahoca and his colleagues applied a comprehensivemethod and obtained interesting findings, we would like to drawattention to some methodological issues concerning the way theusability evaluation methods were employed in this study.The authors mention that Hom [2] identifies three types ofusability evaluation methods, which include testing, inspectionand inquiry and state that they applied heuristic evaluation andcognitive walkthrough (CW), both expert inspection methods, toevaluate the usability of the two prototypes.1. Recruitment of evaluators for heuristic evaluationThe authors recruited potential users of the prototypes as heu-ristic evaluators. Based on a computer literacy test, half of theseusers were classified as novice users. In a study identifying the fac-tors affecting heuristic expertise and levels of expertise permissi-ble to conduct a heuristic evaluation, according to Kirmani [3]three factors (usability experience, experience with the heuristicevaluation, and heuristic training) significantly affect the outcomesof heuristic evaluation. This study showed that domain expertisedoes not have a large impact on the outcomes. Nielsen [4] likewiseshowed that the performance of novice heuristic evaluators, hav-ing general computer knowledge but no special usability expertise,was fairly poor compared to the performance of evaluators withusability expertise. Novice evaluators must first become knowl-edgeable of and proficient in applying heuristics [5]. Therefore,the validity of the heuristic evaluation results of Karahoca et al.using ‘‘usability novices” of which half was computer illiteratecan be disputed.2. Recruitment of evaluators for cognitive walkthroughCW is a usability inspection method that evaluates the easewith which a typical new user can successfully learn to performa task using a given interface design. As also stressed by Hom, inCW either usability specialists or software developers shouldexamine the behavior of the interface [6]. In the study performedby the authors CW was again carried out by end users and notby usability experts. In usability inspection methods such as CW,experts evaluate a user interface without involving users. This isin contrast to usability testing where evaluators let users workwith the system while recording the user sessions for later analysisof usability problems.The approach that the authors followed to assess the learnabil-ity of the prototypes by real users is not in agreement with the CWmethod. In CW, inspectors should know the interface before apply-ing the method and then speculate about the ease with which anovice user can learn how to use the system taking user back-ground knowledge such as computer literacy into account.3. Usability problemsThis usability evaluation study lacks the detection of usabilityproblems,whichisthemaingoalofeveryusabilityevaluationstudy[4,7].Thislackcanaffecttheresultsofthestudyinseveralways.(A)Comparing the effectiveness of the prototypes based on scenariocompletion rates and completion time, without a careful reviewand analysis of the main usability problems that potentially canaffectuserinteractionandtaskoutcomes,doesnotseemvalid.Users,for example, could have completed a scenario in a shorter time byskipping some none mandatory steps hindering them during the

A Framework for Training-Based Estimation in Arbitrarily Correlated Rician MIMO Channels With Rician Disturbance

In this paper, we create a framework for training-based channel estimation under different channel and interference statistics. The minimum mean square error (MMSE) estimator for channel matrix estimation in Rician fading multi-antenna systems is analyzed, and especially the design of mean square error (MSE) minimizing training sequences. By considering Kronecker-structured systems with a combination of noise and interference and arbitrary training sequence length, we collect and generalize several previous results in the framework. We clarify the conditions for achieving the optimal training sequence structure and show when the spatial training power allocation can be solved explicitly. We also prove that spatial correlation improves the estimation performance and establish how it determines the optimal training sequence length. The analytic results for Kronecker-structured systems are used to derive a heuristic training sequence under general unstructured statistics. The MMSE estimator of the squared Frobenius norm of the channel matrix is also derived and shown to provide far better gain estimates than other approaches. It is shown under which conditions training sequences that minimize the non-convex MSE can be derived explicitly or with low complexity. Numerical examples are used to evaluate the performance of the two estimators for different training sequences and system statistics. We also illustrate how the optimal length of the training sequence often can be shorter than the number of transmit antennas.

Introduction of heuristic learning methods in the educational process of higher education

The methods оf heuristic training for increase of educational process' efficiency in high school are analised in this article.

Negotiating Medical Virtues: Toward the Development of a Physician Mistake Disclosure Model

Statistics show that nearly 98,000 patients die each year because of preventable medical mistakes. Despite legal obligations, a majority of physicians either fail to disclose a mistake or disclose it in an incompetent manner, causing detrimental outcomes. This article is the first to synthesize existing research on medical mistakes into an integrative physician mistake disclosure model. The proposed model theorizes that physicians conduct a cost–benefit analysis prior to deciding whether or not to disclose a medical mistake. In the event of disclosure, informational and relational disclosure competence is hypothesized to mediate the inherent detrimental effects of physician defensiveness on immediate and long-term outcomes. The article provides detailed directions for future research and discusses practical implications of the physician mistake disclosure model for physicians and health-care institutions. Most important, the model implies that a supportive organizational climate is needed to curb destructive physician defensiveness, optimize disclosure competence, and minimize detrimental outcomes. Physicians and health-care institutions are advised to collaborate in their attempts to enhance long-term error management and reduce the current number of fatal medical mistakes. The physician mistake disclosure model adds to our current understanding of medical mistake disclosure, and represents a heuristic research and training tool that has the potential to save lives.

An Exploratory Study of College Students' Views of Mathematical Thinking in a Historical Approach Calculus Course

Heuristic training alone is not enough for developing one's mathematical thinking. One missing component is a mathematical point of view. This study reports findings regarding outcomes of a historical approach calculus course to foster Taiwanese college students' views of mathematical thinking. This study consisted of 3 stages. During the initial phase, 44 engineering majors' views on mathematical thinking were tabulated by an open-ended questionnaire, and 9 randomly selected students were invited to participate in follow-up interviews. Students then received an 18-week historical approach calculus course in which mathematical concepts were problematized to challenge their intuition-based empirical beliefs about doing mathematics. Near the end of the semester, all participants answered the identical questionnaire, and we interviewed the same students to pinpoint any shifts in their views on mathematical thinking. We found that participants were more likely to value logical sense, creativity, and imagination in doing mathematics. Further, students were leaning toward a conservative attitude toward certainty of mathematical knowledge. Participants' focus seemingly shifted from mathematics as a product to mathematics as a process.

A heuristic training for support vector regression

A heuristic method for accelerating support vector machine (SVM) training based on a measurement of similarity among samples is presented in this paper. To train SVM, a quadratic function with linear constraints is optimized. The original formulation of the objective function of an SVM is efficient during optimization phase, but the yielded discriminant function often contains redundant terms. The economy of the discriminant function of an SVM is dependent on a sparse subset of the training data, say, selected support vectors by optimization techniques. The motivation for using a sparse controlled version of an SVM is therefore a practical issue since it is the requirement of decreasing computation expense during the SVM testing and enhancing the ability to interpretation of the model. Besides the existing approaches, an intuitive way to achieve this task is to control support vectors sparsely by reducing training data without discounting generalization performance. The most attractive feature of the idea is to make SVM training fast especially for training data of large size because the size of optimization problem can be decreased greatly. In this paper, a heuristic rule is utilized to reduce training data for support vector regression (SVR). At first, all the training data are divided into several groups, and then for each group, some training vectors will be discarded based on the measurement of similarity among samples. The prior reduction process is carried out in the original data space before SVM training, so the extra computation expense may be rarely taken into account. Even considering the preprocessing cost, the total spending time is still less than that for training SVM with the complete training set. As a result, the number of vectors for SVR training becomes small and the training time can be decreased greatly without compromising the generalization capability of SVMs. Simulating results show the effectiveness of the presented method.

Heuristic Training and Performance in Elementary Mathematical Problem Solving

The processes of understanding and solving word problems proceed through the phases of problem translation, problem interpretation, solution planning, solution execution, and solution monitoring. The authors developed a heuristic strategy (SOLVED) to explain these phases in language appropriate to third-, fourth-, and fifth-grade students. Children were trained over several lessons to use it to solve different types of mathematical problems. Results of 2 experiments involving 223 elementary students indicated that SOLVED was more effective in aiding both short-term and delayed problem solving than traditional problem-solving instruction. Accuracy in problem solving was significantly correlated with metacognitive processing. Third-grade students used SOLVED more rapidly and effectively than did older students, and no interaction of learning rate with ability or gender occurred.

An Approach to Robust and Flexible Modelling and Control of pH in Reactors

Preliminary investigations into the potential application of static feed forward neural networks in the dynamic modelling of pH in complex, time-varying systems have been carried out. To assist in network training and testing, a simplified, ‘global first principles’ (FP) model of the pH of such systems was developed, and used successfully to simulate input-output data. Neural networks with input information vectors enhanced by the introduction of auxiliary variables derived from acid-base principles were trained and tested on this data, using both Levenberg-Marquardt (L-M) and heuristic training algorithms. Both algorithms produced good predictions, but the heuristic algorithm required data pre-treatment to minimize its error. However, it trained much faster than the standard, L-M algorithm.

Fuzzy inference using piecewise polynomial interpolation and its application to model-reference fuzzy logic controller

A new fuzzy inference based on piecewise polynomial interpolation similar to the spline technique is proposed. The signed membership function which can encode more information than the usual membership function is also introduced and used together with this new inference method. The fuzzy system using this inference method is more compact compared to other types of fuzzy systems. The computational load when implementing this inference method is almost comparable to the singleton method even if high-order polynomial series are used. A training algorithm similar to the back-propagation algorithm is also proposed to tune these polynomials if numerical training data is available. In contrast to neural network where the trained network is only a function of training data, here, both the heuristic prior knowledge and available training data are used. A simulation example is given to show how this new fuzzy inference can be applied in a model reference closed-loop control system.

Large Tree Classifier with Heuristic Search and Global Training

In the tree classifier with top-down search, a global decision is made via a series of local decisions. Although this approach gains in classification efficiency, it also gives rise to error accumulation which can be very harmful when the number of classes is very large. To overcome this difficulty, a new tree classifier with the following characteristics is proposed: 1) fuzzy logic search is used to find all ``possible correct classes,'' and some similarity measures are used to determine the ``most probable class''; 2) global training is applied to generate extended terminals in order to enhance the recognition rate; 3) both the training and search algorithms have been given a lot of flexibility, to provide tradeoffs between error and rejection rates, and between the recognition rate and speed. A computer simulation of the decision trees for the recognition of 3200 Chinese character categories yielded a very high recognition rate of 99.93 percent and a very high speed of 861 samples/s, when the program was written in a high level language and run on a large multiuser time-sharing computer.

The Formula of Faa Di Bruno

(1980). The Formula of Faa Di Bruno. The American Mathematical Monthly: Vol. 87, No. 10, pp. 805-809.

Explicit Heuristic Training as a Variable in Problem-Solving Performance

This experiment examined the impact that explicit instruction in heuristic strategies, above and beyond problem-solving experience, has on students' problem-solving performance. Two groups of students received training in problem solving, spent the same amount of time working on the same problems, and saw identical problem solutions. But half the students were given a list of five problem-solving strategies and were shown explicitly how the strategies were used. The heuristics group significantly outperformed the other students on posttest problems that were similar to, but not isomorphic to, those used in the problem sets. This lends credence to the idea that explicit instruction in heuristics makes a difference--an idea further supported by the transcripts of students solving the problems out loud.

Explicit Heuristic Training as a Variable in Problem-Solving Performance

Explicit Heuristic Training as a Variable in Problem-Solving Performance