Controlled descent training

In this work, a novel and model-based artificial neural network (ANN) training method is developed supported by optimal control theory. The method augments training labels in order to robustly guarantee training loss convergence and improve training convergence rate. Dynamic label augmentation is proposed within the framework of gradient descent training where the convergence of training loss is controlled. First, we capture the training behavior with the help of empirical Neural Tangent Kernels (NTK) and borrow tools from systems and control theory to analyze both the local and global training dynamics (e.g., stability, reachability). Second, we propose to dynamically alter the gradient descent training mechanism via fictitious labels as control inputs and an optimal state feedback policy. In this way, we enforce locally $$ optimal and convergent training behavior. The novel algorithm, Controlled Descent Training (CDT), guarantees local convergence. CDT unleashes new potentials in the analysis, interpretation, and design of ANN architectures. The applicability of the method is demonstrated on standard regression and classification problems.     

Effects of sign characteristics and training methods on safety sign training effectiveness

This study investigated whether or not training methods affected the effectiveness of symbol training and if there were any relationships between sign symbol characteristics and training effectiveness. Altogether, 26 Mainland China industrial safety signs were used and 60 participants were randomly assigned into four equal-sized groups of control, paired-associate learning, recall training and recognition training. The result was that participants from all the training groups showed significantly greater improvement in comprehension performance than those in the control group, indicating that the training methods improved comprehension of the meaning of safety signs. Participants from the recall training group performed better in the post-training test than those from other training groups. It seems that the recall task elicited a deeper level of learning than the recognition task and that questioning and feedback had a positive effect on training effectiveness. The results also showed that sign characteristics had no significant influence on training effectiveness. It was concluded that recall training is more effective in enhancing comprehension of industrial safety signs than paired-associate learning or recognition training. The findings of this study provide a basis for useful guidelines for designing symbol-training programmes and for designing more user-friendly safety signs.

Statement of Relevance: The present study shows that recall training was more effective in improving comprehension of industrial safety signs than paired-associate learning or recognition training and cognitive sign features did not influence training effectiveness. They provide a basis for useful guidelines for designing symbol-training programmes and for designing more user-friendly safety signs.     

Influences of cognitive style and training method on training effectiveness

The present study compares the relative effects of cognitive style and training method on learners' computer self-efficacy and learning performance by a field experiment. The purpose was to determine which training method could be best utilized in computer-related training while taking trainees' cognitive style into account. The significant two- and three-way interactions indicate the critical roles that personal characteristics and situation factors play as joint determinants of behavior. Gender did significantly moderate the effects of training method on performance and self-efficacy. The cognitive by training method effects were most significant for female participants. This finding suggests that to assist individuals taking IT-related training courses, the contingency effects of gender, cognitive style, training approach, and training objective should be taken into account.     

Perceptual training for visual search

People are better at visual search than the best fully automated methods. Despite this, visual search remains a difficult perceptual task. The goal of this investigation was to experimentally test the ways in which visual search performance could be improved through two categories of training interventions: perceptual training and conceptual training. To determine the effects of each training on a later performance task, the two types of trainings were manipulated using a between-subjects design (conceptual vs. perceptual × training present vs. training absent). Perceptual training led to speed and accuracy improvements in visual search. Issues with the design and administration of the conceptual training limited conclusions on its effectiveness but provided useful lessons for conceptual training design. The results suggest that when the visual search task involves detecting heterogeneous or otherwise unpredictable stimuli, perceptual training can improve visual search performance. Similarly, careful consideration of the performance task and training design is required to evaluate the effectiveness of conceptual training.

Practitioner Summary: Visual search is a difficult, yet critical, task in industries such as baggage screening and radiology. This study investigated the effectiveness of perceptual training for visual search. The results suggest that when visual search involves detecting heterogeneous or otherwise unpredictable stimuli, perceptual training may improve the speed and accuracy of visual search.     

Cutting-plane training of structural SVMs

Discriminative training approaches like structural SVMs have shown much promise for building highly complex and accurate models in areas like natural language processing, protein structure prediction, and information retrieval. However, current training algorithms are computationally expensive or intractable on large datasets. To overcome this bottleneck, this paper explores how cutting-plane methods can provide fast training not only for classification SVMs, but also for structural SVMs. We show that for an equivalent “1-slack” reformulation of the linear SVM training problem, our cutting-plane method has time complexity linear in the number of training examples. In particular, the number of iterations does not depend on the number of training examples, and it is linear in the desired precision and the regularization parameter. Furthermore, we present an extensive empirical evaluation of the method applied to binary classification, multi-class classification, HMM sequence tagging, and CFG parsing. The experiments show that the cutting-plane algorithm is broadly applicable and fast in practice. On large datasets, it is typically several orders of magnitude faster than conventional training methods derived from decomposition methods like SVM-light, or conventional cutting-plane methods. Implementations of our methods are available at .     

The effect of decision-making training: results from a command-and-control training facility

The purpose of this study was to investigate whether training in a microworld had an effect on the decision-making process in a command-and-control training facility. Fourteen battalion commanders participated in the study. Prior to performing the main task in the command-and-control facility, seven participants (the experimental group) received training in the microworld. Seven participants (the control group) received no training before the main task. The results show that the experimental group performed better than the control group, measured by self-ratings and by fulfilling the instructors’ criteria. The experimental group displayed a different decision-making behaviour in the main task than the control group did. The results indicate that the experimental group used some behaviour characteristics they learnt during training, namely “working systematically” and “causal relationship”. Thus, the study shows that favourable behaviour was learnt in the microworld, and that behaviour was subsequently used in the command-and-control training facility.     

Piecewise training for structured prediction

A drawback of structured prediction methods is that parameter estimation requires repeated inference, which is intractable for general structures. In this paper, we present an approximate training algorithm called piecewise training (PW) that divides the factors into tractable subgraphs, which we call pieces, that are trained independently. Piecewise training can be interpreted as approximating the exact likelihood using belief propagation, and different ways of making this interpretation yield different insights into the method. We also present an extension to piecewise training, called piecewise pseudolikelihood (PWPL), designed for when variables have large cardinality. On several real-world natural language processing tasks, piecewise training performs superior to Besag’s pseudolikelihood and sometimes comparably to exact maximum likelihood. In addition, PWPL performs similarly to PW and superior to standard pseudolikelihood, but is five to ten times more computationally efficient than batch maximum likelihood training.     

Nonsmooth training of fuzzy neural networks

The integration of fuzzy methods and neural networks often leads to nonsmoothness of the neural network and, consequently, to a nonsmooth training problem. It is shown, that smooth training methods as e.g. backpropagation fail to converge in this case. Thus a method – based on so called bundle-methods – for training of nonsmooth neural network is presented. Numerical results obtained from a character recognition problem show, that this method still converges where backpropagation fails.     

Comparative study of three training methods for enhancing process control performance: Emphasis shift training,situation awareness training,and drill and practice

Three training methods to improve attention management skills in process control were compared. Forty students from technical disciplines participated in a five-hour module of emphasis shift training (EST), EST combined with situation awareness training (EST/SA), and drill and practice (D&P) on a simulated process control task. Participants were then tested three times for 45 min each (immediately after training, two weeks after training, and six weeks after training) for system control performance and diagnostic performance on familiar and nonfamiliar fault states. D&P led to superior diagnostic performance on familiar system faults. EST/SA training supported the diagnosis of novel system faults. EST was less effective than expected for system control performance. Implications for training design in process control are discussed.     

Fast training of multilayer perceptrons

Training a multilayer perceptron by an error backpropagation algorithm is slow and uncertain. This paper describes a new approach which is much faster and certain than error backpropagation. The proposed approach is based on combined iterative and direct solution methods. In this approach, we use an inverse transformation for linearization of nonlinear output activation functions, direct solution matrix methods for training the weights of the output layer; and gradient descent, the delta rule, and other proposed techniques for training the weights of the hidden layers. The approach has been implemented and tested on many problems. Experimental results, including training times and recognition accuracy, are given. Generally, the approach achieves accuracy as good as or better than perceptrons trained using error backpropagation, and the training process is much faster than the error backpropagation algorithm and also avoids local minima and paralysis.     

Self-regulation training in sport and work

Abstract

This paper describes the current state of the art of self-control or self-regulation training procedures as they are used in sport. At least four important areas of self-control skill and training methods can be identified in the experimental literature as having a beneficial effect upon motor performance. These are goal setting activation control, imagery and attention control. The literature relating to these skills, their pedagogy, and their effects on motor performance is reviewed and conclusions are drawn regarding the implications for sport and work of such skill training In particular, the self-instructional package approach (known as mental training programmes or MTPs) is evaluated as it is the most commonly available form of self-control training.     

Adequacy of training data for evolutionary mining of trading rules

A crucial issue related to data mining on time-series is that of training period duration. The training horizon used impacts the nature of rules obtained and their predictability over time. Longer training horizons are generally sought, in order to discern sustained patterns with robust training data performance that extends well into the predictive period. However, in dynamic environments patterns that persist over time may be unavailable, and shorter-term patterns may hold higher predictive ability, albeit with shorter predictive periods. Such potentially useful shorter-term patterns may be lost when the training duration covers much longer periods. Too short a training duration can, of course, be susceptible to over-fitting to noise. We conduct experiments using different training horizons with daily-data for the S&P500 index and report the sensitivity of the performance of the obtained rules with respect to the training durations. We show that while the performance of the rules in the training period is important for inducing the “best” rules, it is not indicative of their performance in the test-period and propose alternative measures that can be used to help identify the appropriate training durations.     

Effects of three‐dimensional virtual reality and traditional training methods on mental workload and training performance

Industries will implement effective training programs to improve training performance, and an ideal training performance occurs under proper mental workload (MWL). Virtual reality (VR) has recently been widely utilized in training; however, only a few studies have investigated its effects on MWL and training performance simultaneously. The purpose of this study is to investigate the effects of VR training and traditional training methods, such as technical manuals (TM) and multimedia films (MF), on training performance and MWL. The results of the performance measurement show that VR training is considered the best training method compared to TM and MF, particularly in the case of complex tasks. The results of physiological measurements (GSR [galvanic skin response], LF% [low frequency], and LF/HF [high frequency] ratio) show a significant difference between reading TM and using computer (MF and VR), wherein the latter has a lower MWL. However, no significant difference in subjective MWL assessment (NASA‐TLX [task load index]) and HF% measurement is found.     

Effects of sign characteristics and training methods on safety sign training effectiveness

This study investigated whether or not training methods affected the effectiveness of symbol training and if there were any relationships between sign symbol characteristics and training effectiveness. Altogether, 26 Mainland China industrial safety signs were used and 60 participants were randomly assigned into four equal-sized groups of control, paired-associate learning, recall training and recognition training. The result was that participants from all the training groups showed significantly greater improvement in comprehension performance than those in the control group, indicating that the training methods improved comprehension of the meaning of safety signs. Participants from the recall training group performed better in the post-training test than those from other training groups. It seems that the recall task elicited a deeper level of learning than the recognition task and that questioning and feedback had a positive effect on training effectiveness. The results also showed that sign characteristics had no significant influence on training effectiveness. It was concluded that recall training is more effective in enhancing comprehension of industrial safety signs than paired-associate learning or recognition training. The findings of this study provide a basis for useful guidelines for designing symbol-training programmes and for designing more user-friendly safety signs. STATEMENT OF RELEVANCE: The present study shows that recall training was more effective in improving comprehension of industrial safety signs than paired-associate learning or recognition training and cognitive sign features did not influence training effectiveness. They provide a basis for useful guidelines for designing symbol-training programmes and for designing more user-friendly safety signs.     

A personalized learning theory-based cyber-security training exercise

Current enterprises’ needs for skilled cyber-security (CS) professionals have prompted the development of diverse CS training programs and offerings. It has been noted that even though enterprise staff is now more aware of security threats, the number of successful attacks against companies has all but decreased over the years. Several criticisms were raised against current CS training offerings, which often made them inadequate, or unable to change participants’ behavior and security attitude. One of the main factors CS training programs are often not very effective is the lack of engagement or motivation of participants. This is often the result of training not being tailored to the needs or preferences of participants. In our previous work, we tackled this issue by developing a personalized learning theory-based model for developing CS training frameworks. In this work, we utilize the model to develop two CS training exercises: two game-based scenarios using the CS training video game Cyber CIEGE and one table-top team exercise. The exercises are later tested by involving a group of 12 students from the Norwegian Institute of Science and Technology (NTNU) Information Security master’s degree program. According to the results of the experiment and the feedback from the students, students felt more engaged during the exercises due to having been participants in their development process. This has in turn motivated them to continue using the training tools independently in their spare time. Further research is recommended to establish whether the training development model is adequate for different target groups, as well as better performing than other models when developing full-fledged training programs.

     

A fast procedure for the training of neural networks

A new training paradigm for artificial neural networks is described. The technique utilizes a polynomial approximation to the sigmoidal processing function and directly integrates principal components analysis (PCA) into the network training philosophy. A major benefit of the new technique is that off-line network training is ‘one-shot’, contrary to the standard iterative techniques available in the literature. Further training may be performed on-line in a recursive fashion, yielding an adaptive neural network. Additionally, the new philosophy incorporates a systematic procedure for determining the number of neurons in the hidden layer of the network. The training procedure is first described and the implications of the training philosophy discussed. Some results, including applications to industrial chemical processes, are then presented to highlight the power of the technique. The systems considered are a continuous stirred tank reactor and a polymerization reactor.     

Toward a real-time model-based training system

This article describes the development of a real-time model-based training system that provides adaptive “over-the-shoulder” (OTS) instructions to trainees as they learn to perform an Anti-Air Warfare Coordinator (AAWC) task. The long-term goal is to develop a system that will provide real-time instructional materials based on learners’ actions, so that eventually the initial set of instructions on a task can be strengthened, complemented, or overridden at different stages of training. The training system is based on the ACT-R architecture, which serves as the theoretical background for the cognitive model that monitors the learning process of the trainee. An experiment was designed to study the impact of OTS instructions on learning. Results showed that while OTS instructions facilitated short-term learning, (a) they took time away from the processing of current information, (b) their effects tended to decay rapidly in initial stages of training, and (c) their effects on training diminished when the OTS instructions were proceduralized in later stages of training. A cognitive model that learned from both the upfront and OTS instructions was created and provided good fits to the learning and performance data collected from human participants. Our results suggest that to fully capture the symbiotic performance between humans and intelligent training systems, it is important to closely monitor the learning process of the trainee so that instructional interventions can be delivered effectively at different stages of training. We proposed that such a flexible system can be developed based on an adaptive cognitive model that provides real-time predictions on learning and performance.     

Training data reduction to speed up SVM training

Traditional Support Vector Machine (SVM) solution suffers from O(n ²) time complexity, which makes it impractical to very large datasets. To reduce its high computational complexity, several data reduction methods are proposed in previous studies. However, such methods are not effective to extract informative patterns. In this paper, a two-stage informative pattern extraction approach is proposed. The first stage of our approach is data cleaning based on bootstrap sampling. A bundle of weak SVM classifiers are constructed on the sampled datasets. Training data correctly classified by all the weak classifiers are cleaned due to lacking useful information for training. To further extract more informative training data, two informative pattern extraction algorithms are proposed in the second stage. As most training data are eliminated and only the more informative samples remain, the final SVM training time is reduced significantly. Contributions of this paper are three-fold. (1) First, a parallelized bootstrap sampling based method is proposed to clean the initial training data. By doing that, a large number of training data with little information are eliminated. (2) Then, we present two algorithms to effectively extract more informative training data. Both algorithms are based on maximum information entropy according to the empirical misclassification probability of each sample estimated in the first stage. Therefore, training time can be further reduced for training data further reduction. (3) Finally, empirical studies on four large datasets show the effectiveness of our approach in reducing the training data size and the computational cost, compared with the state-of-the-art algorithms, including PEGASOS, LIBLINEAR SVM and RSVM. Meanwhile, the generalization performance of our approach is comparable with baseline methods.     

Agent-customized training for human learning performance enhancement

Training individuals from diverse backgrounds and in changing environments requires customized training approaches that align with the individual learning styles and ever-evolving organizational needs. Scaffolding is a well-established instructional approach that facilitates learning by incrementally removing training aids as the learner progresses. By combining multiple training aids (i.e. multimodal interfaces), a trainer, either human or virtual, must make real-time decisions about which aids to remove throughout the training scenario. A significant problem occurs in implementing scaffolding techniques since the speed and choice of removing training aids must be strongly correlated to the individual traits of a specific trainee. We detail an agent-based infrastructure that supports the customization of scaffolding routines as triggered by the performance of the trainee. The motivation for this agent-based approach is for integration into a training environment that leverages augmented reality (AR) technologies. Initial experiments using the simulated environment have compared the proposed adaptive approach with traditional static training routines. Results show that the proposed approach increases the trainees’ task familiarity and speed with negligible introduction of errors.     

Ophthalmoscopic examination training using virtual reality

Health care professionals perform ophthalmoscopic examinations to detect pathologies of the eye, as well as to evaluate the effects of other diseases, such as high-blood pressure and diabetes. The ophthalmoscopic examination is given using an ophthalmoscope, a hand-held instrument consisting of an adjustable lens and a focused beam of light. The difficulty of the procedure lies in positioning the ophthalmoscope accurately and then correctly identifying the ocular disease symptoms — skills that improve with experience. To improve and accelerate the training of the student, we developed aVirtual Ophthalmoscopic Examination, a three-dimensional real-time computer simulation of the ophthalmoscopic procedure using virtual reality techniques. By navigating and manipulating the virtual ophthalmoscope in the simulation environment, the student learns how to position the instrument properly. Unlike other training aids that use photographic slides to show the full retina, theVirtual Opthalmoscopic Examination programme simulates an accurate view of the retina. By increasing the realism of the training, the transition from the training programme to live examination of patients will become less difficult. The programme was evaluated by graduate nursing students and was shown to be a promising training aid.