Underestimation Problem Research Articles

October 7: The Problem of Underestimation Yet Again

October 7: The Problem of Underestimation Yet Again

End-to-End Autonomous Driving Decision Method Based on Improved TD3 Algorithm in Complex Scenarios.

The ability to make informed decisions in complex scenarios is crucial for intelligent automotive systems. Traditional expert rules and other methods often fall short in complex contexts. Recently, reinforcement learning has garnered significant attention due to its superior decision-making capabilities. However, there exists the phenomenon of inaccurate target network estimation, which limits its decision-making ability in complex scenarios. This paper mainly focuses on the study of the underestimation phenomenon, and proposes an end-to-end autonomous driving decision-making method based on an improved TD3 algorithm. This method employs a forward camera to capture data. By introducing a new critic network to form a triple-critic structure and combining it with the target maximization operation, the underestimation problem in the TD3 algorithm is solved. Subsequently, the multi-timestep averaging method is used to address the policy instability caused by the new single critic. In addition, this paper uses Carla platform to construct multi-vehicle unprotected left turn and congested lane-center driving scenarios and verifies the algorithm. The results demonstrate that our method surpasses baseline DDPG and TD3 algorithms in aspects such as convergence speed, estimation accuracy, and policy stability.

A sound methodology: Measuring experiences of violent conflict through audio self-interviews

This paper investigates the impact of different survey administration methods on the disclosure of sensitive or traumatic experiences. Respondents of a pilot study in Ethiopia were randomly assigned to answer questions either using audio computer-assisted self-interviewing (ACASI) or as part of a face-to-face (FtF) enumerator-based interview. Results indicate that ACASI led to higher disclosure rates of conflict-related experiences, particularly for the most sensitive questions, i.e., when either the respondent or a close friend or family member was the victim, or when the trauma suffered was more severe. ACASI offers a viable solution to measure traumatic conflict-related experience exposure in low-literacy settings, overcoming the underestimation problem commonly observed when using standard survey methods.

Multi-Step Dynamic Ensemble Selection to Estimate Software Effort

ABSTRACT Software Effort Estimation (SEE) is a foremost concern of software companies in order to successfully develop and deliver software products within a defined budget and time. Many software companies fail to deliver the product on time, either due to problems of over-estimation or under-estimation. In order to aid the decision-making process of the analyst and experts, the paper proposed a multi-step dynamic ensemble selection (MS-DES) approach. The DES works in two steps; I) in order to select the suitable models from the pool of models, which are anticipated to perform best when generating a prediction. II) to predict the labeled discretized effort more accurately. The paper utilized four software effort datasets discretized into labeled effort ranges. The performance of the proposed model is evaluated based on the K nearest neighbor oracle (KNORA) canonical approach to DES and in order to reduce the complexity, filter feature selection techniques are applied to extract the relevant feature set. The proposed feature selection-based MS-DES model outplayed the individual models in predicting labeled effort in terms of confusion metrics parameters with an accuracy of more than 90% with all datasets and the results are validated using the ROC curve.

Controlling estimation error in reinforcement learning via Reinforced Operation

In Value-based and Actor-Critic reinforcement learning (RL) methods, both inaccuracy and instability of value estimation will detrimentally affect their performance. Some typical RL methods, such as Maxmin Q-learning and QMD3, are plagued by the underestimation problem while failing to trade off the estimation bias and variance jointly. We propose the Reinforced Operation (RO) to address these shortcomings, which selects the closest median among multiple Q-function. RO is applicable to any model-free RL method. Theoretically, we introduce the Mean Square Error (MSE) to jointly analyze the estimation bias and variance of value estimation methods. We also demonstrate the superiority of RO in MSE reduction and give an upper bound for the estimation bias of value estimation methods with arbitrary distribution to guide the calculation of the estimation bias of value estimation methods. Based on RO, we propose the variants of Q-learning and TD3, Reinforced Q-learning (RQ) and Reinforced Delayed Deep Deterministic policy gradient (RD3), respectively, to tackle different tasks. We empirically demonstrate that our method can reduce estimation error and achieve superior performance on discrete and continuous benchmark tasks.

Tracking changes in chlorophyll-a concentration and turbidity in Nansi Lake using Sentinel-2 imagery: A novel machine learning approach

This study represents the first application of Sentinel-2 remote sensing imagery and model fusion techniques to assess the chlorophyll-a (Chla) concentration and turbidity in Nansi Lake, Shandong Province, China, from 2016 to 2022. First, we innovatively employed the stacking method to fuse eight fundamentally different Machine Learning (ML) models, each utilising 20 and 17 feature bands, resulting in the development of a robust algorithm for estimating the Chla concentration and turbidity in Nansi Lake. The results demonstrate that the Stacking Model has achieved outstanding theoretical generalisation capability. Second, the sensitivity of the model to extreme value data in the sample was quantified, and we found that compared with extreme gradient boosting (XGBoost), the optimal performance of the Stacking Model improved by 12%, to some extent, it solved the problem of high-value underestimation and low-value overestimation. The SHapley Additive exPlanations (SHAP) results revealed that features such as Three Bands, Enhanced Three, Rrs492/Rrs560, Rrs705/Rrs665 play a crucial role in estimating Chla concentration. For the turbidity estimation, the Normalized Difference Turbidity Index (NDTI), Rrs705+Rrs560, Rrs865-Rrs740 made significant contributions. Finally, we utilised the Stacking Model to create spatiotemporal maps of the Chla concentration and turbidity in Nansi Lake from 2016 to 2022. We analysed the causes of the water quality changes and explored the driving factors. Compared with previous studies, this paper provides a new idea for the monitoring of lake water quality parameters by using the high resolution of Sentinel-2 image and the high precision of model fusion technology, these results can provide a reference for similar water area research and decision-making support for environment-related departments.

Secondary Carnitine Deficiency in Neonates and Infants Requiring Surgery for Intestinal Obstructions-An Underestimated and Undermanaged Problem.

This study aims to elucidate the relationship between intestinal obstruction and carnitine deficiency in neonates and infants. We retrospectively reviewed medical records of 330 neonates and infants, younger than six months, who underwent surgery for intestinal obstruction at our institute from January 2009 to April 2022. The analysis focused on clinical symptoms, related signs, complications, and etiology of the intestinal obstruction. Tandem mass spectrometry (MS/MS) or urine organic acid analysis was conducted for 47 patients, revealing carnitine deficiency in 16 patients. The incidence of carnitine deficiency was 34.0% in the suspicious group and 4.8% overall, significantly higher than in the general population in Taiwan. Notably, patients with carnitine deficiency experienced prolonged ileus, with a mean fasting duration of 41.7 days (range 7.8-65.5 days), compared to 10.8 days (range 8.2-13.4 days) in patients without carnitine deficiency. Carnitine replacement therapy was administered to twelve patients at dosages ranging from 32 to 90 mg/kg/day. One patient exhibited a drug allergy with skin rashes. Our findings suggest that carnitine deficiency should be considered in cases of neonatal and infant intestinal obstruction. Replacement therapy is straightforward and can be prognostically beneficial for some patients. Therefore, we recommend generalizing MS/MS and urine organic acid analysis, particularly for patients with prolonged ileus.

Magnetic and thermoelectric properties of REMoN3 (RE = La, Ce, Pr, Nd, and Sm): A first‐principles study

AbstractWe investigated the magnetic and thermoelectric properties of REMoN3 (RE = La, Ce, Pr, Nd, Sm) perovskites using the full potential linearized augmented plane wave (FP‐LAPW) method. To overcome the problem of underestimation of electronic interaction, we employed the DFT + U approach to accurately map the electronic structure of these compounds. Our study shows an increasing trend in the magnetic moments with the increasing number of unpaired electrons in RE. Among these compounds, SmMoN3 possesses a large magnetic moment, which is suitable for applications such as memory devices and sensors. Interestingly, all these perovskites display ferromagnetic behavior except CeMoN3, which exhibits an antiferromagnetic nature. Furthermore, our analysis indicates n‐type thermoelectric behavior in all these materials. The compound, namely PrMoN3, exhibits a high figure of merit among REMoN3, which can be improved by modifying the lattice sites.

Eating Event Recognition Using Accelerometer, Gyroscope, Piezoelectric, and Lung Volume Sensors.

In overcoming the worldwide problem of overweight and obesity, automatic dietary monitoring (ADM) is introduced as support in dieting practises. ADM aims to automatically, continuously, and objectively measure dimensions of food intake in a free-living environment. This could simplify the food registration process, thereby overcoming frequent memory, underestimation, and overestimation problems. In this study, an eating event detection sensor system was developed comprising a smartwatch worn on the wrist containing an accelerometer and gyroscope for eating gesture detection, a piezoelectric sensor worn on the jaw for chewing detection, and a respiratory inductance plethysmographic sensor consisting of two belts worn around the chest and abdomen for food swallowing detection. These sensors were combined to determine to what extent a combination of sensors focusing on different steps of the dietary cycle can improve eating event classification results. Six subjects participated in an experiment in a controlled setting consisting of both eating and non-eating events. Features were computed for each sensing measure to train a support vector machine model. This resulted in F1-scores of 0.82 for eating gestures, 0.94 for chewing food, and 0.58 for swallowing food.

Data-Driven Outage Restoration Time Prediction via Transfer Learning With Cluster Ensembles

This paper develops a data-driven approach to accurately predict the restoration time of outages under different scales and factors. To achieve the goal, the proposed method consists of three stages. First, given the unprecedented amount of data collected by utilities, a sparse dictionary-based ensemble spectral clustering (SDESC) method is proposed to decompose historical outage datasets, which enjoys good computational efficiency and scalability. Specifically, each outage sample is represented by a linear combination of a small number of selected dictionary samples using a density-based method. Then, the dictionary-based representation is utilized to perform the spectral analysis to group the data samples with similar features into the same subsets. In the second stage, a knowledge-transfer-added restoration time prediction model is trained for each subset by combining weather information and outage-related features. The transfer learning technology is introduced to deal with the underestimation problem caused by data imbalance in different subsets, thus improving the model performance. Furthermore, to connect unseen outages with the learned outage subsets, a t-distributed stochastic neighbor embedding-based strategy is applied. The proposed method fully builds on and is also tested on a large real-world outage dataset from a utility provider with a time span of six consecutive years. The numerical results validate that our method has high prediction accuracy while showing good stability against real-world data limitations.

Toward a positive compensation policy for rail transport via mechanism design: The case of China Railway Express

This paper addresses one of the key challenges in the China Railway Express (CR Express) service industry, which is the asymmetry of information between the government and a CR Express operator. We formulate the CR Express service delivery problem with asymmetric information and other unique characteristics as the incentive compensation problem via the principal-agent approach. This is the first paper to develop an incentive-compatible (IC) compensation mechanism to improve CR Express service in the presence of adverse selection and moral hazard. Compared to the generic form of compensation, this paper proves that such an IC scheme with a linear compensation contract is optimal. Our optimal compensation mechanism provides a “positive incentive”, where the monetary compensation is higher for the CR Express operator with lower operating costs, as opposed to the current compensation policy, which can create a “negative incentive”. We find both underestimation and underinvestment problems, which become smaller when the CR Express operator's gross margin is relatively high. When the CR Express service level is discrete-valued, it is optimal to adopt a fixed compensation scheme. Finally, in the scenario with multiple CR Express operators, we propose a generalized Vickrey (GV) auction, which implicitly obtains both true gross margin and service level. The compensation offered by the government is indeed negative in the case of GV auction. Surprisingly, in this scenario, we find problems of both underestimation and overestimation.

Instance-level loss based multiple-instance learning framework for acoustic scene classification

An acoustic scene is inferred by detecting properties combining diverse sounds and acoustic environments. This study is intended to discover these properties effectively using multiple-instance learning (MIL). MIL, also known as a weakly supervised learning approach, is a strategy for extracting an instance vector from an audio chunk that composes an audio clip and utilizing these unlabeled instances to infer a scene corresponding to the input data. However, many studies pointed out an underestimation problem of MIL. In this study, we propose an enhanced MIL framework more suitable for ASC systems by defining instance-level labels and loss to extract and cluster instances effectively. Furthermore, we design a lightweight convolutional neural network named FUSE comprising frequency-, temporal-sided depthwise, and pointwise convolutional filters. Experimental results show that the confidence and proportion of positive instances significantly increase compared to vanilla MIL, overcoming the underestimation problem and improving the classification accuracy even higher than the supervised learning. The proposed system achieved a performance of 81.1%, 72.3%, and 58.3% on the TAU urban acoustic scenes 2019, 2020 mobile, and 2022 mobile datasets with 139 K parameters, respectively. In particular, it achieves the highest performance among the systems having under the 1 M parameters for the TAU urban acoustic scenes 2019 dataset.

Monitoring of subpixel impervious surface dynamics using seasonal time series Landsat 8 OLI imagery

Since impervious surface is one of the core urban ecological indices, its high-precision time series monitoring at regional or global scales is highly desirable. Spectral features, spatial features, and temporal features have been successfully applied to regional impervious surface monitoring. However, few studies focus on the extraction of optimal features for subpixel impervious surface monitoring at the regional or global scales. In order to solve the above problem, this study proposes a new method for subpixel time series impervious surface estimation based on optimal spectral-temporal features. The proposed method first extracts various spectral indices from time series Landsat imagery to construct a spectral-temporal feature library. A feature scoring model is then established to optimize spectral-temporal features from the feature library. Based on the optimal spectral-temporal features, a modified semi-supervised Fuzzy C-means (FCM) method is proposed to identify stable pixels and extract endmembers from stable pixels. Finally, the endmembers are applied to the Fully Constrained Least Squares (FCLS) for time series impervious surface abundance extraction. The optimal spectral-temporal features can effectively distinguish impervious surfaces from pervious surfaces through feature optimization selection, and improve the feature utilization performances, thereby obtaining high-precision impervious surface mapping. The proposed method is applied in Wuhan, China, from 2013 to 2021, using seasonal time series Landsat 8 OLI images. Experimental results show that the impervious surface area of Wuhan expands from 567.27 km2 in winter 2013 to 1056.95 km2 in winter 2021 at the pixel scale. The impervious surface area expands from 545.80 km2 in winter 2013 to 921.01 km2 in winter 2021 at the subpixel scale. The subpixel impervious surface monitoring effectively solve the overestimation and underestimation problem of the pixel-scale impervious surface monitoring. This study characterizes the long-term dynamics of impervious surfaces at the subpixel scale and can provide reliable urban growth information for evaluating the impact of urbanization on the ecological environment.

Contemporary ceramic art of China: a look from the perspective of the art market

The object of this research is contemporary Chinese art in the context of the latest trends in the development of the global and national art market. The subject of the study is the problem of underestimation of modern ceramic art in China and its potential from the standpoint of the art market. The history of the issue is considered, the analysis of the main directions of research development in this subject area is carried out. In addition, the article clarifies the existing concept of "modern ceramic art" in China. Special attention is paid to the analysis of the most significant modern objects of Chinese ceramics presented in the domestic art market of the country. The features of the artistic originality of the masters Li Jiusheng, Wang Silyan, Zhang Songmao, Guo Aihe are revealed. The scientific novelty of the research is connected with the development of ideas about the modern ceramic art of China, its role in the modern art market, both at the national and international levels. For the first time, the criteria for the demand for modern porcelain are revealed; the idea is stated that the problem of collecting modern Chinese ceramics remains on the periphery of scientific discourse. A special contribution is the establishment and justification of the artistic value of a number of works carried out for the first time in science, which were sold at various auctions for fabulous money. The main conclusions of the study are the following. Currently, the development of the art market indicates that experts are beginning to realize the artistic value of modern Chinese ceramics art. The greatest interest among collectors is aroused by works in the spirit of "ceramic painting" with the use of modern technologies that allow expanding the expressive possibilities of painting on porcelain. The boom in collecting modern Chinese porcelain, observed since the beginning of the XXI century, testifies to the high level of artistic execution of objects of modern ceramic art in China, as well as such qualities as innovation and non-standardness in terms of choosing themes, visual effects, expression of the author's idea, rethinking academic traditions.

Improved General Polarimetric Model-Based Decomposition for Coherency Matrix

A representative general polarimetric model-based decomposition framework was proposed by Chen et al., which implements a simultaneous full-parameter inversion by using complete polarimetric information and solves several limitations in previous decomposition methods. However, there are still shortcomings in Chen’s work. Firstly, only the real part of the parameter β in the generalized surface scattering model is considered. Secondly, inappropriate initial input values may lead to local optima in the nonlinear least squares optimization algorithm. Thirdly, the volume scattering component is underestimated in the volume scattering-dominated scene, but overestimated in buildings with large orientation (LOB) areas. Finally, nonlinear optimization is time-consuming computationally. To overcome those issues, an improved generalized polarimetric model-based decomposition method is proposed in this paper. The imaginary part of the parameter β is incorporated into the decomposition framework of the proposed method. Ingeniously utilizing the internal relationship in the generic equations composed of coherent matrix elements, the model parameters can be inversed by simplifying the nonlinear equations to linear equations. Therefore, compared with Chen’s method, the proposed method does not rely on the initial input values, and improves the computational efficiency. In addition, a hierarchical decomposition scheme is presented to solve the problem of underestimation or overestimation of volume scattering component mentioned above. The performance and advantages of this method are evaluated with L-band and C-band polarimetric synthetic aperture radar (PolSAR) data sets. Comparison studies are carried out with other model-based decomposition methods, demonstrating that the proposed method can further improve decomposition performance, especially in LOB areas.

Estimating tropical cyclone intensity using dynamic balance convolutional neural network from satellite imagery

Accurate estimation of tropical cyclone (TC) intensity helps to understand the evolution of TCs throughout their life cycle and plays an essential role in mitigating TC impact. Although TC intensity estimation methods based on deep learning have made significant progress, the existing techniques do not apply good methods to overcome the intensity overestimation and underestimation problems caused by the unbalanced distribution of TC data. Therefore, we propose a dynamic balance convolutional neural network to overcome these issues. The model consists of two branches, one branch is the learning of the raw data, and the other is the learning of strong (weak) TCs that account for a few data samples. Finally, the model is dynamically adjusted by adaptive trade-off parameters, gradually from the learning of the raw data to the learning of strong (weak) TCs, thus reducing errors in underestimation (overestimation) of strong (weak) TCs. Furthermore, an attention mechanism is employed to obtain the correlation between channels to improve the accuracy of TC intensity estimation further. We used globally 1285 TC cases from 2003–2016 to train the model and globally 94 TC cases from 2017 as independent test data. The results showed that the root-mean-square error of TC intensity estimation was 8.32 kt, 35% lower than that of the advanced Dvorak technique and 26% lower than that of the deep learning method visual geometry group (VGG). For a subset of 482 samples (from East Pacific and Atlantic) analyzed with reconnaissance observations, a root-mean-square intensity difference of 7.95 kt is achieved. Finally, we explored the model’s feature learning process and the contribution of each component of the satellite image to the TC intensity estimation through model visualization.

A cross-sectoral analysis of energy shortages in Pakistan: based on supply-driven input-output model

The impacts of energy shortages characterized by regular blackouts, natural gas, and electricity load shedding in Pakistan affected each economic sector, causing an energy-induced crisis and ecological sustainability issues. This study was conducted to reveal the benefits of renewable energy and describe the economic losses associated with electricity unavailability using supply-driven input-output as a price model across 34 sectors. The results revealed that exogenous shocks in electricity prices are responsible for bringing significant fluctuations across the business cycle in the country. Similarly, the overall output of Pakistan’s economy will decrease by 24.89 rupees due to a 1-kilowatt-hour reduction in electricity supply. Moreover, both forward and backward linkages of Pakistan’s economy revealed that higher electricity allocation coefficients pose significant output impacts on most sectors. We conclude that indirect output impacts require due consideration to avoid the underestimation problem due to total electricity shortages. It is recommended that the government provide a social and legal framework to boost the environmental sustainability and economic activities in the textile, oil refining, production of cement, and fertilizer sectors for sustainable economic growth.

Research on Flexible Resource Dynamic Interactive Regulation Technology for Microgrids with High Permeable New Energy

The efficiency of on-site consumption of new energy and the economy of dispatching strategy for that in modern microgrids are increasingly concerning, which are closely related to the microgrid control model with source-load uncertainty. To this end, this paper proposes the multiagent hierarchical IQ (λ)-HDQC regulation strategy to realize the source-load-storage-charging collaborative control of the microgrid model with high-permeable new energy. The first layer adopts the IQ (λ) strategy, which avoids the overestimation and underestimation problems of traditional reinforcement learning by the coupled estimation method. The second layer adopts the HDQC allocation strategy, which solves the problem of low utilization of new energy in the proportional allocation method and improves the adaptability of the regulation strategy in the complex stochastic environment. The interaction of the two-layer strategies realizes the source-network-load-storage-charge global dynamic interactive regulation of microgrids. Indicators of energy efficiency are constructed in this paper to measure the simulation results. And the superiority of the proposed strategy is verified through the simulation results of the microgrid system.

Bayesian Analysis of Multivariate Matched Proportions with Sparse Response

Multivariate matched proportions (MMP) data appear in a variety of contexts including post-market surveillance of adverse events in pharmaceuticals, disease classification, and agreement between care providers. It consists of multiple sets of paired binary measurements taken on the same subject. While recent work proposes methods to address the complexities of MMP data, the issue of sparse response, where no or very few “yes” responses are recorded for one or more sets, is unaddressed. The presence of sparse response sets results in the underestimation of variance components, loss of coverage, and lowered power in existing methods. Bayesian methods, which have not previously been considered for MMP data, provide a useful framework when sparse responses are present. In particular, the Bayesian probit model in combination with mean model prior specifications provides an elegant solution to the problem of variance underestimation. We examine a multivariate probit-based approach using hierarchical horseshoe-like priors along with a Bayesian functional principal component analysis (FPCA) to model the latent covariance. We show that our approach performs well on MMP data with sparse responses and outperforms existing methods. In a re-examination of a study on the system of care (SOC) framework for children with mental and behavioral disorders, we are able to provide a more complete picture of the relationships in the data. Our analysis provides additional insights into the functioning on the SOC that a previous univariate analysis missed.

Gearbox fault diagnosis based on generalized multivariate logarithmic regularization

Feature extraction based on sparse representation is widely applied in the field of mechanical fault diagnosis. L1 norm regularization is a classical sparse regularization method, but this method has sparse underestimation for large-value features. A signal sparse representation method based on generalized multivariate logarithmic regularization is proposed in this paper. First, the sparse penalty term in the proposed method is designed according to the minimum convolution and logarithmic function, namely the generalized logarithmic non-convex penalty function. Then, the convexity condition of the objective function is studied to verify the feasibility of the method. The applicability of the method is also improved by analyzing the parameter constraint relation in the objective function. Finally, the sparse optimal solution is obtained by the forward-back splitting algorithm. Experiments show that compared with other non-convex sparse models, the proposed method can solve the problem of sparse underestimation more effectively and improve the reliability of gearbox fault diagnosis.