Penalty Level Research Articles

Tennis Technology Recognition and Training Attitude Analysis Based on Artificial Intelligence Sensor

In order to solve the problem of traditional tennis serving technique teaching, subjective, and experience-based teaching methods, the author proposes a method to extract tennis training movements based on artificial intelligence sensor video analysis. This method requires the use of ordinary cameras and computer technology; video analysis technology is used to guide tennis teaching and training. The result obtained is as follows: in the experimental group and the control group, at the end of the teaching stage, the movement characteristics of each link of the body are quite different, which can reach 15%-20%; when the experimental group and the control group did the “scratching back” movement during the teaching period, at the end of teaching, the students in the experimental group were close to high-level tennis players in their movement skills, and the movement gap was only about 5%. It is proved that the method proposed by the author provides an objective and scientific basis for measuring the referee’s penalty level. In the grass-roots tennis teaching and training, it has important value and significance and has a great promotion prospect.

An efficient GPU-parallel coordinate descent algorithm for sparse precision matrix estimation via scaled lasso

The sparse precision matrix plays an essential role in the Gaussian graphical model since a zero off-diagonal element indicates conditional independence of the corresponding two variables given others. In the Gaussian graphical model, many methods have been proposed, and their theoretical properties are given as well. Among these, the sparse precision matrix estimation via scaled lasso (SPMESL) has an attractive feature in which the penalty level is automatically set to achieve the optimal convergence rate under the sparsity and invertibility conditions. Conversely, other methods need to be used in searching for the optimal tuning parameter. Despite such an advantage, the SPMESL has not been widely used due to its expensive computational cost. In this paper, we develop a GPU-parallel coordinate descent (CD) algorithm for the SPMESL and numerically show that the proposed algorithm is much faster than the least angle regression (LARS) tailored to the SPMESL. Several comprehensive numerical studies are conducted to investigate the scalability of the proposed algorithm and the estimation performance of the SPMESL. The results show that the SPMESL has the lowest false discovery rate for all cases and the best performance in the case where the level of the sparsity of the columns is high.

The influence of online review adoption on the profitability of capacitated supply chains

The paper explores the influence of online review adoption on supply chain profitability under the presence of a capacity constraint. Nowadays, customers increasingly rely on online reviews for decision making, and online retailers regard reviews as a norm. Although online reviews have been extensively examined in marketing disciplines, little research has been conducted to investigate their influence from a supply chain perspective. In addition, previous research has largely focused on how online review information can influence customer purchase behaviours, but ignores the more basic decision: whether and when companies should adopt reviews. This paper examines the online review adoption decision from a capacitated supply chain perspective through mathematical modelling and simulation. The simulation considers the influence of variables including online review adoption decision, capacity constraint level, lost sales penalty level, and product quality estimation on supply chain profitability. Generally, we find that online reviews can bring more profit to the supply chain than without online reviews, although such influence is moderated by the other three variables. The findings reveal the complexity of the contextual variable impacts on online review adoption, and demonstrate that decisions concerning the adoption of online reviews should take all supply-chain-related variables into consideration rather than only aiming for increasing customer orders.

LASSO-driven inference in time and space

We consider the estimation and inference in a system of high-dimensional regression equations allowing for temporal and cross-sectional dependency in covariates and error processes, covering rather general forms of weak temporal dependence. A sequence of regressions with many regressors using LASSO (Least Absolute Shrinkage and Selection Operator) is applied for variable selection purpose, and an overall penalty level is carefully chosen by a block multiplier bootstrap procedure to account for multiplicity of the equations and dependencies in the data. Correspondingly, oracle properties with a jointly selected tuning parameter are derived. We further provide high-quality de-biased simultaneous inference on the many target parameters of the system. We provide bootstrap consistency results of the test procedure, which are based on a general Bahadur representation for the Z-estimators with dependent data. Simulations demonstrate good performance of the proposed inference procedure. Finally, we apply the method to quantify spillover effects of textual sentiment indices in a financial market and to test the connectedness among sectors.

Risk-aware service level agreement modeling in smart grid

As advanced Smart Grid environments grow from a simple grid towards a complex provider ecosystem, there is an uncertain challenge on those grid environments that need to manage a risk paradigm. We present an automated risk-aware service level agreements modeling to the grid provider for speed automated pricing and getting better performance to program as an agent-oriented platform. The key idea of our novel approach is proposing a risk level agreements contract and a pricing model to decrease the complexity of previous methods from an off-line service level agreement to an on-line risk level agreement for managing the risk lifecycle of contracts used to record the rights and obligations of the services and their consumers. Based on a risk level agreements contract the model optimizes resource management according to the business objective level of the provider with an online risk-aware rendezvous to define the penalty level of the cost model. The corresponding quality of service criteria is defined based on multi-class risk-aware service level agreements between Smart Grid providers and their power consumers which include the tail distributions of the per-class costs in addition to the more standard quality of service metrics such as throughput and mean delays. Our empirical experiments show the benefits of the proposed approach.

An experimental study on the sale of counterfeit products under monitoring policies

This study examines the decision behaviour of a seller who may sell either a genuine product or a deceptive counterfeit product under a monitoring policy characterised by an inspection frequency and a penalty level. A unique feature of selling deceptive counterfeit products is that the seller incurs significant losses when the deceptive counterfeit product is inspected, but enjoys significant gains if it is not. We conduct a laboratory experiment to investigate the seller’s behaviour regarding his/her selling strategy. In contrast to the standard theoretical predictions, we observe that the seller’s strategy exhibits a “deviate-to-equal-probability” effect, a diminishing marginal impact of the penalty level, and dynamic behavioural biases such as the “gambler’s fallacy” and speculation. A behavioural model is developed using bounded rationality and prospect theory to describe the observed behaviours, including the perception of the inspection probability in an inverse “S-shaped” pattern. Our results imply that, when designing effective monitoring policies, policymakers need to discreetly choose inspection frequencies depending on the penalty levels and to take into account sellers’ dynamic behavioural biases.

Sorted concave penalized regression

The Lasso is biased. Concave penalized least squares estimation (PLSE) takes advantage of signal strength to reduce this bias, leading to sharper error bounds in prediction, coefficient estimation and variable selection. For prediction and estimation, the bias of the Lasso can be also reduced by taking a smaller penalty level than what selection consistency requires, but such smaller penalty level depends on the sparsity of the true coefficient vector. The sorted $\ell_{1}$ penalized estimation (Slope) was proposed for adaptation to such smaller penalty levels. However, the advantages of concave PLSE and Slope do not subsume each other. We propose sorted concave penalized estimation to combine the advantages of concave and sorted penalizations. We prove that sorted concave penalties adaptively choose the smaller penalty level and at the same time benefits from signal strength, especially when a significant proportion of signals are stronger than the corresponding adaptively selected penalty levels. A local convex approximation for sorted concave penalties, which extends the local linear and quadratic approximations for separable concave penalties, is developed to facilitate the computation of sorted concave PLSE and proven to possess desired prediction and estimation error bounds. Our analysis of prediction and estimation errors requires the restricted eigenvalue condition on the design, not beyond, and provides selection consistency under a required minimum signal strength condition in addition. Thus, our results also sharpens existing results on concave PLSE by removing the upper sparse eigenvalue component of the sparse Riesz condition.

AIC for the group Lasso in generalized linear models

When covariates are assumed to be clustered in groups in regression problems, the group Lasso is useful, because it tends to drive all the weights in one group to zero together. The group Lasso includes a tuning parameter which controls a penalty level, and an unbiased estimator of the true prediction error is derived as a $$C_p$$ -type criterion to select an appropriate value of the tuning parameter. However, in general, the $$C_p$$ -type criterion cannot be derived in generalized linear regressions such as a logistic regression. Hence, this paper obtains the AIC for the group Lasso based on its original definition under the framework of generalized linear models. In terms of computational cost, our criterion is clearly better than the cross validation, but it is shown through simulation studies that the performance of the our criterion is almost the same as or better than that of the cross validation.

Fiber Integrated Wavelength Converter Based on a Silicon Core Fiber With a Nano-Spike Coupler

An all-fiber integrated nonlinear silicon photonic wavelength converter has been proposed and fabricated using the silicon core fiber platform. The silicon fiber was spliced directly to a conventional single mode fiber, facilitated via an inverse tapered nano-spike that helped to reduce the mode mismatch between the different core materials. Four-wave mixing-based wavelength conversion with an efficiency as high as −22.1 dB has been achieved for selected wavelengths across the C-band in a device length of only ~ 1 cm. Successful conversion of quadrature phase-shift keying signals at a 20-Gb/s bitrate, with a 1 to 2 dB penalty level at the bit error ratio $\text {(BER)} =3.8\times 10^{-3}$ , was used to demonstrate the suitability of the silicon fiber device for the construction of ultra-compact, all-fiber-based optical signal processing systems.

The impact of liner shipping unreliability on the production–distribution scheduling of a decentralized manufacturing system

Liner shipping uncertainty affects global supply chains reliability seriously. The pressure from downstream forces the make-to-order manufacturer to face high penalty cost due to stochastic shipping lead-times. A new risk-averse model is proposed aimed at improving production scheduling reliability integrated with shipment assignments in terms of total operating cost. The closed form of the risk cost of job-shipment assignment under given reliability levels is formulated, which verifies the effects of shipping uncertainty on the production scheduling and operating cost. The computational and statistical evaluation demonstrate our approach can compensate the amplification effects of the high penalty level and shipping uncertainty.

AIC for the non-concave penalized likelihood method

Non-concave penalized maximum likelihood methods are widely used because they are more efficient than the Lasso. They include a tuning parameter which controls a penalty level, and several information criteria have been developed for selecting it. While these criteria assure the model selection consistency, they have a problem in that there are no appropriate rules for choosing one from the class of information criteria satisfying such a preferred asymptotic property. In this paper, we derive an information criterion based on the original definition of the AIC by considering minimization of the prediction error rather than model selection consistency. Concretely speaking, we derive a function of the score statistic that is asymptotically equivalent to the non-concave penalized maximum likelihood estimator and then provide an estimator of the Kullback–Leibler divergence between the true distribution and the estimated distribution based on the function, whose bias converges in mean to zero.

Studi Tentang Teori GONE dan Pengaruhnya Terhadap Fraud Dengan Idealisme Pimpinan Sebagai Variabel Moderasi: Studi Pada Pengadaan Barang/Jasa di Pemerintahan

This study aims to examine the effect of GONE theory’s variables on the procurement fraud of government goods/services by using variables of leadership idealism as a moderating variable. This study uses question instruments for each variable: Greed, Opportunity, Need, and Exposes from Bologne (1993), Leadership Idealism from Tappen (1994) and Procurement Fraud of Government Goods/Services. This study collects primary data gathered from the respondent’s perception in Kebumen District Government agencies. The analysis technique used is Moderated Regression Analysis (MRA). The results of this study demonstrate that the existence of Exposes (low penalty level) and lack of leadership idealism will increase the potential procurement fraud of goods/services in government. In other words without the punishment that can provide a deterrent effect and less idealistic leadership will be more and more potential individuals within the bureaucracy associated in the procurement process of goods/services to conduct fraud.

LASSO-Driven Inference in Time and Space

We consider the estimation and inference in a system of high-dimensional regression equations allowing for temporal and cross-sectional dependency in covariates and error processes, covering rather general forms of weak dependence. A sequence of large-scale regressions with LASSO is applied to reduce the dimensionality, and an overall penalty level is carefully chosen by a block multiplier bootstrap procedure to account for multiplicity of the equations and dependencies in the data. Correspondingly, oracle properties with a jointly selected tuning parameter are derived. We further provide high-quality de-biased simultaneous inference on the many target parameters of the system. We provide bootstrap consistency results of the test procedure, which are based on a general Bahadur representation for the Z-estimators with dependent data. Simulations demonstrate good performance of the proposed inference procedure. Finally, we apply the method to quantify spillover effects of textual sentiment indices in a financial market and to test the connectedness among sectors.

Design and Implementation of Driving Time Regulation

Our object of study is welfare optimal driving time regulations in professional road transportation. Due to fatigue, traffic accident risks are supposed to increase as driving times rise. Conversely, the quantity and quality of road infrastructure affect productivity and safety in transportation positively. As the typical driver does not bear all the social costs when accidents happen, in the absence of public regulation, she has an incentive to drive too many hours. Hence, we present two types of public regulatory tools: a uniform driving time restriction and a uniform tax. We then compare the likely outcomes of these regulations (second-best policies) with the welfare optimal (first-best) solution. Moreover, as driving time restrictions are commonly applied worldwide, we study the problem of implementing such prescriptions. When public authorities choose optimal resources in driving time restrictions, the detection of the flouting of these restrictions and the penalty levels for non-compliance, the welfare gains involved must balance the direct and indirect enforcement costs. For example, it follows that the welfare-optimal penalty level should not be so high that the most efficient companies are always forced to comply with the uniform driving time restriction.

A Penalized Likelihood Method for Structural Equation Modeling.

A penalized likelihood (PL) method for structural equation modeling (SEM) was proposed as a methodology for exploring the underlying relations among both observed and latent variables. Compared to the usual likelihood method, PL includes a penalty term to control the complexity of the hypothesized model. When the penalty level is appropriately chosen, the PL can yield an SEM model that balances the model goodness-of-fit and model complexity. In addition, the PL results in a sparse estimate that enhances the interpretability of the final model. The proposed method is especially useful when limited substantive knowledge is available for model specifications. The PL method can be also understood as a methodology that links the traditional SEM to the exploratory SEM (Asparouhov & Muthén in Struct Equ Model Multidiscipl J 16:397-438, 2009). An expectation-conditional maximization algorithm was developed to maximize the PL criterion. The asymptotic properties of the proposed PL were also derived. The performance of PL was evaluated through a numerical experiment, and two real data illustrations were presented to demonstrate its utility in psychological research.

Crime and punishment in classroom: a game-theoretic approach for student cheating

We provide the microeconomic foundations of cheating in classroom through a static game with complete information. Our setting is composed by two students, who must choose whether or not to cheat, and a professor, who must choose how much effort to exert in trying to catch dishonest students. Our findings support the determinants of cheating found by the empirical literature, mainly those related to the penalty's level. It is also emphasized the importance of professors being well-motivated (with low disutility of effort) and worried about fairness in classroom. The two extensions of the baseline model reinforce the importance of the cost-benefit analysis to understand dishonest behavior in classroom.

Ultra-Compact Amorphous Silicon Waveguide for Wavelength Conversion

In this letter, we demonstrate, for the first time, successful four wave mixing (FWM)-based wavelength conversion of binary phase shift keyed (BPSK) and quadrature phase shift keyed (QPSK) signals, at 20-Gb/s bitrate, in a 1-mm long amorphous silicon waveguide. A maximum FWM-efficiency of −26 dB was achieved by employing a pump power of just 70 mW, establishing this technology as a contender for the development of ultra-compact, low power, silicon photonics wavelength converter. Bit error ratio measurements demonstrated successful conversion with less than 1 dB penalty level, for both BPSK and QPSK signals (at $\textrm {BER} = 10^{-5}$ ).

Safety Impact of Enforcement, Penalty Level, and Fuel Prices

Safety Impact of Enforcement, Penalty Level, and Fuel Prices

Subjective evaluation of loudness and preference for noise containing audible tones

Noise in living environments often contains unwanted tonal components that contribute to the unpleasant and intrusive character of a sound and usually lead to an increased annoyance. In the measurement and assessment of noise immissions after the German DIN 45681:2005-3 standard, the higher annoyance due to audible tonal components is covered by so called tone adjustments. Depending on the SNR of a tonal component within each critical band, the level of a sound can be charged with a penalty of up to 6 dB. In this study, the psychoacoustic penalty levels for sounds containing tonal components are directly determined in listening tests. The points of subjective equality (PSEs) for loudness and preference are determined using a matching procedure. The level of the test sound with tonal components is varied until it is equally loud / preferred as a tone-free reference sound that is constant in level. The level difference between test and reference sound at the PSEs is a direct measure of the penalty level. In listening tests with pure multi-tone sounds the penalty levels were found to be well above 6 dB, indicating that for these stimuli the German DIN standard is underestimating the penalty level.

Incorporating Grouping Information in Bayesian Variable Selection with Applications in Genomics

In many applications it is of interest to determine a limited number of important explanatory factors (representing groups of potentially overlapping predictors) rather than original predictor variables. The often imposed require-ment that the clustered predictors should enter the model simultaneously may be limiting as not all the variables within a group need to be associated with the out-come. Within-group sparsity is often desirable as well. Here we propose a Bayesian variable selection method, which uses the grouping information as a means of in-troducing more equal competition to enter the model within the groups rather than as a source of strict regularization constraints. This is achieved within the context of Bayesian LASSO (least absolute shrinkage and selection operator) by allowing each regression coefficient to be penalized differentially and by considering an additional regression layer to relate individual penalty parameters to a group identification matrix. The proposed hierarchical model therefore enables inference simultaneously on two levels: (1) the regression layer for the continuous outcome in relation to the predictors and (2) the regression layer for the penalty param-eters in relation to the grouping information. Both situations with overlapping and non-overlapping groups are applicable. The method does not assume within-group homogeneity across the regression coefficients, which is implicit in many structured penalized likelihood approaches. The smoothness here is enforced at the penalty level rather than within the regression coefficients. To enhance the potential of the proposed method we develop two rapid computational procedures based on the expectation maximization (EM) algorithm, which offer substantial time savings in applications where the high-dimensionality renders Markov chain Monte Carlo (MCMC) approaches less practical. We demonstrate the usefulness of our method in predicting time to death in glioblastoma patients using pathways of genes.