Pyramid Bayesian Method for Model Uncertainty Evaluation of Semantic Segmentation in Autonomous Driving

With the fast-developing deep learning models in the field of autonomous driving, the research on the uncertainty estimation of deep learning models has also prevailed. Herein, a pyramid Bayesian deep learning method is proposed for the model uncertainty evaluation of semantic segmentation. Semantic segmentation is one of the most important perception problems in understanding visual scene, which is critical for autonomous driving. This study to optimize Bayesian SegNet for uncertainty evaluation. This paper first simplifies the network structure of Bayesian SegNet by reducing the number of MC-Dropout layer and then introduces the pyramid pooling module to improve the performance of Bayesian SegNet. mIoU and mPAvPU are used as evaluation matrics to test the proposed method on the public Cityscapes dataset. The experimental results show that the proposed method improves the sampling effect of the Bayesian SegNet, shortens the sampling time, and improves the network performance.

SDTM: A Novel Topic Model Framework for Syndrome Differentiation in Traditional Chinese Medicine

Syndrome differentiation is the most basic diagnostic method in traditional Chinese medicine (TCM). The process of syndrome differentiation is difficult and challenging due to its complexity, diversity, and vagueness. Recently, artificial intelligent methods have been introduced to discover the regularities of syndrome differentiation from TCM medical records, but the existing DM algorithms failed to consider how a syndrome is generated according to TCM theories. In this paper, we propose a novel topic model framework named syndrome differentiation topic model (SDTM) to dynamically characterize the process of syndrome differentiation. The SDTM framework utilizes latent Dirichlet allocation (LDA) to discover the latent semantic relationship between symptoms and syndromes in mass of Chinese medical records. We also use similarity measurement method to make the uninterpretable topics correspond with the labeled syndromes. Finally, Bayesian method is used in the final differentiated syndromes. Experimental results show the superiority of SDTM over existing topic models for the task of syndrome differentiation.

The Effect of Test Length on the Accuracy of Estimating Ability Parameter in the Two- and Three-Parameter Logistic Models: Comparison by Using the Bayesian Method of Expected Prior Mode and Maximum Likelihood Estimation

This study aims to compare the effect of test length on the degree of ability parameter estimation in the two-parameter and three-parameter logistic models, using the Bayesian method of expected prior mode and maximum likelihood. The experimental approach is followed, using the Monte Carlo method of simulation. The study population consists of all subjects with the specified ability level. The study includes random samples of subjects and of items. Results reveal that estimation accuracy of the ability parameter in the two-parameter logistic model according to the maximum likelihood method and the Bayesian method increases with the increase in the number of test items. Results also show that with long and average length tests, the effectiveness is related to the maximum likelihood method and to all conditions of the sample size, whereas in short tests, the Bayesian method of prior mode outperformed in all conditions. Results indicate that the increase of the ability parameter in the three-parameter logistic model increases with the increase of test items number. The Bayesian method outperforms with respect to the accuracy of estimation at all conditions of the sample size, whereas in long tests the maximum likelihood method outperforms at all different conditions. Received: 17 September 2021 / Accepted: 24 November 2021 / Published: 3 January 2022

ANALISIS METODE BAYESIAN PADA SISTEM ANTREAN RAWAT JALAN DI RSUP Dr. KARIADI DENGAN DISTRIBUSI SAMPEL POISSON DAN GEOMETRIK

Hospital service facilities cannot be separated from queuing events. Queues are an unavoidable part of life, but they can be minimized with a good system. The purpose of this study was to find out how the queuing system at Dr. Kariadi. Bayesian method is used to combine previous research and this research in order to obtain new information. The sample distribution and prior distribution obtained from previous studies are combined with the sample likelihood function to obtain a posterior distribution. After calculating the posterior distribution, it was found that the queuing model in the outpatient installation at Dr. Kariadi Semarang is (G/G/c): (GD/∞/∞) where each polyclinic has met steady state conditions and the level of busyness is greater than the unemployment rate so that the queuing system at Dr. Kariadi is categorized as good, except in internal medicine poly.

PEMODELAN SISTEM ANTREAN PELAYANAN BUS JALUR BARAT TERMINAL TIRTONADI KOTA SURAKARTA DENGAN METODE BAYESIAN

Tirtonadi is the largest bus station in Surakarta City. The departure line is devided into two lines, namely west line and east line. The west line serves buses to the west of Surakarta City. The number of buses that enter and leave the station every day causes bus queues. Modeling the queue system and analyzing the system performance measure aims to determine wether the bus service system is good or not. The queue system model is obtained by finding the distribution of arrival patterns and service patterns using the Bayesian method. This method is used because it combines the information from the current research and the prior information from the previous research. The queueing condition of the five lanes in the west line meets steady state conditions because the utility value is less than 1. The queue displant is First Come First Service (FCFS) with unlimited customers and unlimited calling sources. Based on the posterior distribution, the queue system of service bus is (GAMM/IG/1):(GD/∞/∞) for Solo-Jakarta-Bandung lane and Pedesaan lane, while for Solo-Purwokerto-Cilacap, Solo-Yogyakarta, and Solo-Semarang has the queue system (GAMM/GAMM/1):(GD/∞/∞). The queue system of service bus for each lane has good services based on the value of system performance measure.

ANALISIS METODE BAYESIAN MENGGUNAKAN NON-INFORMATIF PRIOR UNIFORM DISKRIT PADA SISTEM ANTREAN PELAYANAN GERBANG TOL MUKTIHARJO

The growth rate of the traffic that is high resulting in congestion on the road network system. One of the government's efforts in addressing the issue with the build highways to reduce congestion, especially in large cities. One of the queuing phenomena that often occurs in the city of Semarang is the queue at the Toll Gate Muktiharjo, that the queue of vehicles coming to make toll payment. This study aims to determine how the service system at the Toll Gate Muktiharjo. This can be known by getting a queue system model and a measure of system performance from the distribution of arrival and service. The distribution of arrival and service are determined by finding the posterior distribution using the Bayesian method. The bayesian method combine the likelihood function of the sample and the prior distribution. The likelihood function is a negative binomial. The prior distribution used a uniform discrete. Based on the calculations and analysis, it can be concluded that the queueing system model at the Toll Gate Muktiharjo is a (Beta/Beta/5):(GD/∞/∞). The queue simulation obtained that the service system Toll Gate Muktiharjo is optimal based on the size of the system performance because busy probability is higher than jobless probability.

Forward-Looking Super-Resolution Imaging for Sea-Surface Target with Multi-Prior Bayesian Method

Traditional forward-looking super-resolution methods mainly concentrate on enhancing the resolution with ground clutter or no clutter scenes. However, sea clutter exists in the sea-surface target imaging, as well as ground clutter when the imaging scene is a seacoast.Meanwhile, restoring the contour information of the target has an important effect, for example, in the autonomous landing on a ship. This paper aims to realize the forward-looking imaging of a sea-surface target. In this paper, a multi-prior Bayesian method, which considers the environment and fuses the contour information and the sparsity of the sea-surface target, is proposed. Firstly, due to the imaging environment in which more than one kind of clutter exists, we introduce the Gaussian mixture model (GMM) as the prior information to describe the interference of the clutter and noise. Secondly, we fuse the total variation (TV) prior and Laplace prior, and propose a multi-prior to model the contour information and sparsity of the target. Third, we introduce the latent variable to simplify the logarithm likelihood function. Finally, to solve the optimal parameters, the maximum posterior-expectation maximization (MAP-EM) method is utilized. Experimental results illustrate that the multi-prior Bayesian method can enhance the azimuth resolution, and preserve the contour information of the sea-surface target.

TreeKnit: Inferring Ancestral Reassortment Graphs of influenza viruses

When two influenza viruses co-infect the same cell, they can exchange genome segments in a process known as reassortment. Reassortment is an important source of genetic diversity and is known to have been involved in the emergence of most pandemic influenza strains. However, because of the difficulty in identifying reassortments events from viral sequence data, little is known about its role in the evolution of the seasonal influenza viruses. Here we introduce TreeKnit, a method that infers ancestral reassortment graphs (ARG) from two segment trees. It is based on topological differences between trees, and proceeds in a greedy fashion by finding regions that are compatible in the two trees. Using simulated genealogies with reassortments, we show that TreeKnit performs well in a wide range of settings and that it is as accurate as a more principled bayesian method, while being orders of magnitude faster. Finally, we show that it is possible to use the inferred ARG to better resolve segment trees and to construct more informative visualizations of reassortments.