Variational Bayesian-Based Improved Maximum Mixture Correntropy Kalman Filter for Non-Gaussian Noise

The maximum correntropy Kalman filter (MCKF) is an effective algorithm that was proposed to solve the non-Gaussian filtering problem for linear systems. Compared with the original Kalman filter (KF), the MCKF is a sub-optimal filter with Gaussian correntropy objective function, which has been demonstrated to have excellent robustness to non-Gaussian noise. However, the performance of MCKF is affected by its kernel bandwidth parameter, and a constant kernel bandwidth may lead to severe accuracy degradation in non-stationary noises. In order to solve this problem, the mixture correntropy method is further explored in this work, and an improved maximum mixture correntropy KF (IMMCKF) is proposed. By derivation, the random variables that obey Beta-Bernoulli distribution are taken as intermediate parameters, and a new hierarchical Gaussian state-space model was established. Finally, the unknown mixing probability and state estimation vector at each moment are inferred via a variational Bayesian approach, which provides an effective solution to improve the applicability of MCKFs in non-stationary noises. Performance evaluations demonstrate that the proposed filter significantly improves the existing MCKFs in non-stationary noises.

Super-Human or Sub-Human? Positive Dehumanization and Gender in Public Performance Evaluations

Dehumanizing language, language that compares human beings to animals or machines, is typically thought of in problematic cases, where it is designed to denigrate individuals or entire groups in society. But, this language can also be used to praise—describing an employee as a machine can be done to signify super-human characteristics. We find that positive dehumanizing language has no effect on evaluations of a public employee’s competence, but do have an effect on evaluations of warmth. Contrary to expectations, we find no differences in these effects based on the gender of the employee.

Employee onboarding and satisfaction in US manufacturing companies

Purpose The purpose of this study was to identify primary issues related to employee onboarding and satisfaction in US-based manufacturing companies. Design/methodology/approach In total, 19 focus groups using semi-structured interviews with senior management, middle management, tenured employees and new employees were conducted with personnel from five manufacturing companies located in the Midwest USA. Onboarding procedures, training manuals, employee satisfaction surveys and performance evaluations were subsequently reviewed. Findings Insufficient onboarding, poor communication and a perceived lack of support were reported as satisfaction concerns by manufacturing employees. In addition, management had vastly differing perspectives regarding the work environment when responses were contrasted with those from new or tenured employees. Originality/value This paper reveals contributing factors that influence satisfaction early and throughout an employee’s tenure with small- to medium-sized US manufacturing companies.

Performance Evaluation of an Autonomously Driven Agricultural Vehicle in an Orchard Environment

To address the problems of inefficient agricultural production and labor shortages, there has been active research to develop autonomously driven agricultural machines, using advanced sensors and ICT technology. Autonomously driven speed sprayers can also reduce accidents such as the pesticide poisoning of farmers, and vehicle overturn that frequently occur during spraying work in orchards. To develop a commercial, autonomously driven speed sprayer, we developed a prototype of an autonomously driven agricultural vehicle, and conducted performance evaluations in an orchard environment. A prototype of the agricultural vehicle was created using a rubber-tracked vehicle equipped with two AC motors. A prototype of the autonomous driving hardware consisted of a GNSS module, a motion sensor, an embedded board, and an LTE module, and it was made for less than $1000. Additional software, including a sensor fusion algorithm for positioning and a path-tracking algorithm for autonomous driving, were implemented. Then, the performance of the autonomous driving agricultural vehicle was evaluated based on two trajectories in an apple farm. The results of the field test determined the RMS, and the maximums of the path-following errors were 0.10 m, 0.34 m, respectively.

Stable Support for Democracy in East and Southeast Asia? Examining Citizens’ Trust in Democratic Institutions

Recent democratic regressions and crises suggest democracy is at risk across East and Southeast Asia. One of the factors that can determine democratic stability are citizens’ attitudes. While previous research has concentrated on support for democracy-in-principle, this contribution argues that it is political trust, i.e. support for democracy-in-practice, which is crucial for democratic stability. For democracies to be stable, political trust should be high as well as rooted in long-term factors like liberal democratic value orientations or social trust to protect it from short-term fluctuations following economic crises or political scandals. This contribution therefore examines not only the current levels and development of political trust but also whether it is influenced more by long-term factors (liberal democratic value orientations, social trust) or short-term factors (economic performance evaluations, incumbent support). The empirical analysis shows political trust in five East and Southeast Asian democracies (Indonesia, Japan, Mongolia, South Korea, Taiwan) to be mostly mediocre and primarily dependent on economic performance evaluations and incumbent support. Among the five democracies, citizens in Japan appear most resilient to democratic regressions; on the other hand, Taiwanese democracy seems least equipped to master future crises.

Discrete Wavelet Transforms-Based Analysis of Accelerometer Signals for Continuous Human Cardiac Monitoring

Measuring cardiac activity from the chest using an accelerometer is commonly referred to as seismocardiography. Unfortunately, it cannot provide clinically valid data because it is easily corrupted by motion artefacts. This paper proposes two methods to improve peak detection from noisy seismocardiography data. They rely on discrete wavelet transform analysis using either biorthogonal 3.9 or reverse biorthogonal 3.9. The first method involves slicing chest vibrations for each cardiac activity, and then detecting the peak location, whereas the other method aims at detecting the peak directly from chest vibrations without segmentation. Performance evaluations were conducted on signals recorded from small children and adults based on missing and additional peaks. Both algorithms showed a low error rate (15.4% and 2.1% for children/infants and adults, respectively) for signals obtained in resting state. The average error for sitting and breathing tasks (adults only) was 14.4%. In summary, the first algorithm proved more promising for further exploration.

ProteinGLUE: A multi-task benchmark suite for self-supervised protein modeling.

Self-supervised language modeling is a rapidly developing approach for the analysis of protein sequence data. However, work in this area is heterogeneous and diverse, making comparison of models and methods difficult. Moreover, models are often evaluated only on one or two downstream tasks, making it unclear whether the models capture generally useful properties. We introduce the ProteinGLUE benchmark for the evaluation of protein representations: a set of seven tasks for evaluating learned protein representations. We also offer reference code, and we provide two baseline models with hyperparameters specifically trained for these benchmarks. Pre-training was done on two tasks, masked symbol prediction and next sentence prediction. We show that pre-training yields higher performance on a variety of downstream tasks such as secondary structure and protein interaction interface prediction, compared to no pre-training. However, the larger base model does not outperform the smaller medium. We expect the ProteinGLUE benchmark dataset introduced here, together with the two baseline pre-trained models and their performance evaluations, to be of great value to the field of protein sequence-based property prediction. Availability: code and datasets from https://github.com/ibivu/protein-glue

An Improved Backoff Scheme and Its Performance Analysis for Full Duplex MAC Protocols in VLC Networks

IEEE 802.15.7 Visible Light Communication (VLC) networks suffer from performance degradation caused by the hidden device collisions due to the directional transmission with narrow beamwidth. One of the solutions for mitigating the hidden device collisions is to employ a full-duplex transmission technique. As a side effect of the full-duplex transmission in the VLC networks, however, the data-packet discard due to the retransmission limitation occurs frequently in the networks. This paper proposes an improved backoff scheme and its performance analysis to suppress the packet discard. The proposed backoff scheme increases the Backoff Exponent (BE) and the Number of Backoff stage (NB) in IEEE 802.15.7 only when the data packet transmission fails. To evaluate the system performance theoretically, this paper also provides the Markov-chain model for channel access with the proposed scheme. The performance evaluations through simulation and theoretical analysis show the effectiveness of the proposed scheme.