Bayesian Disturbance Injection: Robust imitation learning of flexible policies for robot manipulation.

Humans demonstrate a variety of interesting behavioral characteristics when performing tasks, such as selecting between seemingly equivalent optimal actions, performing recovery actions when deviating from the optimal trajectory, or moderating actions in response to sensed risks. However, imitation learning, which attempts to teach robots to perform these same tasks from observations of human demonstrations, often fails to capture such behavior. Specifically, commonly used learning algorithms embody inherent contradictions between the learning assumptions (e.g., single optimal action) and actual human behavior (e.g., multiple optimal actions), thereby limiting robot generalizability, applicability, and demonstration feasibility. To address this, this paper proposes designing imitation learning algorithms with a focus on utilizing human behavioral characteristics, thereby embodying principles for capturing and exploiting actual demonstrator behavioral characteristics. This paper presents the first imitation learning framework, Bayesian Disturbance Injection (BDI), that typifies human behavioral characteristics by incorporating model flexibility, robustification, and risk sensitivity. Bayesian inference is used to learn flexible non-parametric multi-action policies, while simultaneously robustifying policies by injecting risk-sensitive disturbances to induce human recovery action and ensuring demonstration feasibility. Our method is evaluated through risk-sensitive simulations and real-robot experiments (e.g., table-sweep task, shaft-reach task and shaft-insertion task) using the UR5e 6-DOF robotic arm, to demonstrate the improved characterization of behavior. Results show significant improvement in task performance, through improved flexibility, robustness as well as demonstration feasibility.

Analysis of characteristics of non-suicidal self-injury in 120 adolescents / 上海预防医学

ObjectiveTo explore the characteristics of adolescent non-suicidal self-injury （NSSI） behavior， and to enable clinicians to further understand adolescents with NSSI behavior， so as to make better clinical diagnosis and intervention. MethodsFrom July 2022 to June 2023， 120 adolescent patients with NSSI behavior were selected from the outpatient department of our hospital by convenience sampling， and the general demographic data were collected by self-made general situation questionnaire. The characteristics and motivation of NSSI among adolescents with NSSI were analyzed by using the NSSI questionnaire and Ottawa self-injury inventory （OSI）. ResultsThe average age of the first NSSI in the adolescents was 12.90±1.233， and the most common was 13 years old. The most common site of NSSI was the lower arm or wrist （58.33%）， followed by the hand （27.5%）. The most commonly used NSSI was "intentionally cutting oneself" （68.33%）； The most common motivation for NSSI is emotional regulation. ConclusionThere are significant differences in gender among adolescents with NSSI behavior. The age of first NSSI is concentrated， and the highest incidence is in the 11‒14 years old. The common way of self-injury is cutting. NSSI is usually associated with the intention to relieve suffering and is characterized by high frequency， using variety of methods， and low mortality. Adolescents with NSSI often choose to implement NSSI with emotion regulation as the main function， including external emotion regulation and internal emotion regulation.

Criminological Characteristics and Criminal Responsibility Capacity of Offenders with Alcoholism / 中山大学学报(医学科学版)

ObjectiveTo analyze the criminal behavior characteristics of alcohol-related crime offenders and explore the relationship between criminal behavior characteristics and criminal responsibility capacity. MethodsBasic information， alcohol use information， psychiatric symptoms before and after the crime， criminological behavioral characteristics and conclusion of judicial evaluation were collected. ResultsThe drinking history of the chronic alcoholism group 25.0 （13.3~30.0） years was significantly longer than that of the acute alcoholism group 1.2 （0~14.3） years. In the chronic alcoholism group， 85.0% drank alcohol at least once a day， 52.5% had morning or bedtime drinking habits， and 92.5% drank mainly alone. Violent crimes accounted for 57.6%. Delirium existed in 52.5% of the chronic alcoholics' mental state at the time of the crime， and 84.6% of the acute alcoholics' mental state was hazy. In the chronic alcoholism group， 42.5% committed the crime with pathological motive， and in the acute alcoholism group， 69.2% committed the crime with realistic motive. Acute alcoholism group 96.2% were assessed as complete criminal responsibility capacity， and chronic alcoholism group 50.0% were assessed as complete criminal responsibility capacity. ConclusionCompared with acute alcoholism offenders， chronic alcoholics had a longer drinking duration， more frequency drinking， and obvious morning drinking or bedtime drinking habits. The nature of alcoholism crime cases was mostly violent. Compared with the chronic alcoholism group， the acute alcoholism group had highly selective motives for committing crimes and were mostly rated as complete criminally responsibility.

Ship Classification and Anomaly Detection Based on Spaceborne AIS Data Considering Behavior Characteristics.

With the establishment of satellite constellations equipped with ship automatic identification system (AIS) receivers, the amount of AIS data is continuously increasing, and AIS data have become an important part of ocean big data. To further improve the ability to use AIS data for maritime surveillance, it is necessary to explore a ship classification and anomaly detection method suitable for spaceborne AIS data. Therefore, this paper proposes a ship classification and anomaly detection method based on machine learning that considers ship behavior characteristics for spaceborne AIS data. In view of the characteristics of different types of ships, this paper introduces the extraction and analysis of ship behavior characteristics in addition to traditional geometric features and discusses the ability of the proposed method for ship classification and anomaly detection. The experimental results show that the classification accuracy of the five types of ships can reach 92.70%, and the system can achieve better results in the other classification evaluation metrics by considering the ship behavior characteristics. In addition, this method can accurately detect anomalous ships, which further proves the effectiveness and feasibility of the proposed method.

An Effective Gaze-Based Authentication Method with the Spatiotemporal Feature of Eye Movement.

Eye movement has become a new behavioral feature for biometric authentication. In the eye movement-based authentication methods that use temporal features and artificial design features, the required duration of eye movement recordings are too long to be applied. Therefore, this study aims at using eye movement recordings with shorter duration to realize authentication. And we give out a reasonable eye movement recording duration that should be less than 12 s, referring to the changing pattern of the deviation degree between the gaze point and the stimulus point on the screen. In this study, the temporal motion features of the gaze points and the spatial distribution features of the saccade are using to represent the personal identity. Two datasets are constructed for the experiments, including 5 s and 12 s of eye movement recordings. On the datasets constructed in this paper, the open-set authentication results show that the Equal Error Rate of our proposed methods can reach 10.62% when recording duration is 12 s and 12.48% when recording duration is 5 s. The closed-set authentication results show that the Equal Error Rate of our proposed methods can reach 5.25% when recording duration is 12 s and 7.82% when recording duration is 5 s. It demonstrates that the proposed method provides a reference for the eye movements data-based identity authentication.

Effect of Five Driver's Behavior Characteristics on Car-Following Safety.

Driver's behavior characteristics (DBCs) influence car-following safety. Therefore, this paper aimed to analyze the effect of different DBCs on the car-following safety based on the desired safety margin (DSM) car-following model, which includes five DBC parameters. Based on the Monte Carlo simulation method, the effect of DBCs on car-following safety is investigated under a given rear-end collision (RECs) condition. We find that larger subjective risk perception levels can reduce RECs, a smaller acceleration sensitivity (or a larger deceleration sensitivity) can improve car-following safety, and a faster reaction ability of the driver can avoid RECs in the car-following process. It implies that DBCs would cause a traffic wave in the car-following process. Therefore, a reasonable value of DBCs can enhance traffic flow stability, and a traffic control strategy can improve car-following safety by using the adjustment of DBCs.

Evaluation of Walking Comfort in Children's School Travel at Street Scale: A Case Study in Tianjin (China).

(1) Background: school travel is an important part of a child's daily activities. A comfortable walking environment can encourage children to walk to school. The existing methods of evaluating walking environments are not specific to children's walks to school. (2) Methods: this study proposes a method of evaluating walking comfort in children traveling to school at street scale. Related indexes were selected that reflect children's school travel behavior and their needs in street environments based on walking environment audit tools. Factor analysis was then used to calculate the relative weight of each index. (3) Results: the new evaluation method was tested in the neighborhoods around the First Central Primary School in Hedong District, Tianjin, China. The walking comfort for children's school travel was evaluated in eight indexes: effective street width; street flatness; street cleanliness; interface diversity; buffer; shade coverage; green looking ratio; and sound decibels. Different classes and types of streets were found to have various vulnerabilities. (4) Conclusions: this evaluation method can accurately locate the weak spots in streets to improve the local policymakers' perception of street environments, which can greatly facilitate the implementation of precise measures to promote children walking to school.

Study of hot spring bathing and chronic disease prevalence among residents between 30 and 65 in typical hot spring areas of Guizhou Province / 上海预防医学

Objective To investigate hot spring bathing behavior and chronic disease prevalence of residents aged 30 to 65 in typical hot spring areas of Guizhou Province. Methods Totally 3 708 individuals between 30 and 65 in five typical hot spring areas of Guizhou Province were included in this study. Their hot spring bathing behavior and chronic disease prevalence were recorded. Unconditional logistic regression was used to analyze the association between hot spring bathing and the incidence of chronic diseases. Results 71.3% of the residents in the hot spring areas had bathed in hot springs, with the proportion higher in males than females(χ2=92.856, P < 0.001). 76.3% of people took hot spring baths fewer than or equal to 6 times per year, 30.8% of residents often chose to take hot spring baths in the evening, and 37.6% of the residents took hot spring baths for 21-40 minutes. Bathing was more common in winter. 11.5% and 17.9% of the residents added other substances and used hot spring facilities, and most of them added traditional Chinese medicine and used the sauna. 46.6% and 41.6 % of the residents took hot spring baths to relax and relieve fatigue. The self-reported prevalence rate of chronic diseases was 522.1‰, and the top five single diseases were lumbar intervertebral disc disease(119.7‰), hypertension(118.1‰), urinary calculus(77.7‰), cervical intervertebral disc disease(53.7‰)and osteogenesis(44.5‰). Compared with those who never bathed in hot springs, taking hot spring baths 7-11 times a year was negatively correlated with chronic diseases.(OR=0.551, 95%CI: 0.345-0.879). Conclusion Hot spring bathing is popular in typical hot spring areas of Guizhou Province, but its frequency is not high. The overall prevalence of chronic diseases is relatively high, and the order of most prevalent chronic diseases is different from that in other areas. Proper hot spring bathing is negatively correlated with chronic diseases.

A Comparative Study of Accident Risk Related to Speech-Based and Handheld Texting during a Sudden Braking Event in Urban Road Environments.

The use of mobile phones while driving is a very common phenomenon that has become one of the main causes of traffic accidents. Many studies on the effects of mobile phone use on accident risk have focused on conversation and texting; however, few studies have directly compared the impacts of speech-based texting and handheld texting on accident risk, especially during sudden braking events. This study aims to statistically model and quantify the effects of potential factors on accident risk associated with a sudden braking event in terms of the driving behavior characteristics of young drivers, the behavior of the lead vehicle (LV), and mobile phone distraction tasks (i.e., both speech-based and handheld texting). For this purpose, a total of fifty-five licensed young drivers completed a driving simulator experiment in a Chinese urban road environment under five driving conditions: baseline (no phone use), simple speech-based texting, complex speech-based texting, simple handheld texting, and complex handheld texting. Generalized linear mixed models were developed for the brake reaction time and rear-end accident probability during the sudden braking events. The results showed that handheld texting tasks led to a delayed response to the sudden braking events as compared to the baseline. However, speech-based texting tasks did not slow down the response. Moreover, drivers responded faster when the initial time headway was shorter, when the initial speed was higher, or when the LV deceleration rate was greater. The rear-end accident probability respectively increased by 2.41 and 2.77 times in the presence of simple and complex handheld texting while driving. Surprisingly, the effects of speech-based texting tasks were not significant, but the accident risk increased if drivers drove the vehicle with a shorter initial time headway or a higher LV deceleration rate. In summary, these findings suggest that the effects of mobile phone distraction tasks, driving behavior characteristics, and the behavior of the LV should be taken into consideration when developing algorithms for forward collision warning systems.

[Applicability and feasibility of "Lingnanzhun" -an "Internet Plus-based HIV Self-testing Tool" targeting MSM in Guangzhou].

Objective: This article was to evaluate the applicability and feasibility of "Lingnanzhun" -an "Internet Plus-based HIV Self-testing Tool" targeting MSM in Guangzhou. Hopefully, the results could be used to improve the existing HIV testing services and to support the implementation and scale-up of HIV self-testing programs. Methods: Data were collected from a survey on HIV testing preferences among the Internet-using MSM in April to June, 2015. Univariate and multivariate logistic regression analysis were applied to identify factors associated with the use of HIV self-testing service provided by Lingnanzhun. Information related to the users of Lingnanzhun during September 2014 and December 2018 was also collected. Results: 769 MSM were recruited as participants. Of them, age distribution was 16-77(28.6±6.8) years old, 88.3%(679/769) were unmarried, 42.2%(325/769) were registered residents of Guangzhou, 82.1%(631/769) had university or college education. Among them, 195 (25.4%) used the HIV self- testing program of Lingnanzhun while 574 (74.6%) using the clinic service. Compared with the clinic service users, the Lingnanzhun users showed the following characteristics: longer experience in the MSM community ≥10 years (32.8% vs. 20.9%, 64/195 vs. 120/574); having male casual sexual partners ≥2 (42.1% vs. 29.6%, 82/195 vs. 170/574); having group sex (6.2% vs. 2.6%, 12/195 vs. 15/574) and having commercial sex with men (13.8% vs. 3.0%, 27/195 vs. 17/574). Data from the Lingnanzhun users showed that a total of 3 000 users had practised 5 038 times of self HIV-testings. 11.4% (343/3 000) of the Lingnanzhun users had never been tested. Conclusions: It was applicable and feasible to provide Internet-based HIV self-testing service to MSM as a significant complement to the traditional facility-based HIV testing services. It was also useful in accessing those who were having higher risk or had never received HIV testing so as to increase the testing uptake and the frequency.

Applicability and feasibility of "Lingnanzhun" -an "Internet Plus-based HIV Self-testing Tool" targeting MSM in Guangzhou / 中华流行病学杂志

Objective@#This article was to evaluate the applicability and feasibility of "Lingnanzhun" -an "Internet Plus-based HIV Self-testing Tool" targeting MSM in Guangzhou. Hopefully, the results could be used to improve the existing HIV testing services and to support the implementation and scale-up of HIV self-testing programs.@*Methods@#Data were collected from a survey on HIV testing preferences among the Internet-using MSM in April to June, 2015. Univariate and multivariate logistic regression analysis were applied to identify factors associated with the use of HIV self-testing service provided by Lingnanzhun. Information related to the users of Lingnanzhun during September 2014 and December 2018 was also collected.@*Results@#769 MSM were recruited as participants. Of them, age distribution was 16-77(28.6±6.8) years old, 88.3%(679/769) were unmarried, 42.2%(325/769) were registered residents of Guangzhou, 82.1%(631/769) had university or college education. Among them, 195 (25.4%) used the HIV self- testing program of Lingnanzhun while 574 (74.6%) using the clinic service. Compared with the clinic service users, the Lingnanzhun users showed the following characteristics: longer experience in the MSM community ≥10 years (32.8% vs. 20.9%, 64/195 vs. 120/574); having male casual sexual partners ≥2 (42.1% vs. 29.6%, 82/195 vs. 170/574); having group sex (6.2% vs. 2.6%, 12/195 vs. 15/574) and having commercial sex with men (13.8% vs. 3.0%, 27/195 vs. 17/574). Data from the Lingnanzhun users showed that a total of 3 000 users had practised 5 038 times of self HIV-testings. 11.4% (343/3 000) of the Lingnanzhun users had never been tested.@*Conclusions@#It was applicable and feasible to provide Internet-based HIV self-testing service to MSM as a significant complement to the traditional facility-based HIV testing services. It was also useful in accessing those who were having higher risk or had never received HIV testing so as to increase the testing uptake and the frequency.

Characteristics of sexual partner network and high-risk sex behaviors among men who have sex with men / 上海预防医学

[Objective]To ascertain characteristics of sexual partner network and high-risk sex behaviors among men who have sex with men (MSM) , providing references for intervention of high risk behavior in MSM. [Methods] Face-to-face questionnaire interview was conducted among MSM recruited by voluntary counseling and testing clinics, peer mobilization, internet mobilization, outreach intervention, and so on.Descriptive statistics, χ2test, logistic regression analysis were applied for statistical analysis. [Results]A total of 334 MSM was investigated. Over the past 6 months, proportion of female partners, fixed male partners, multiple male partners, occasional and business male partners was 37.13％, 43.11％, 52.69％, 56.29％, and 4.19％ respectively. Logistic regression analysis showed that respondents who were married, less educated, non-Shanghai domicile, and non-homosexual had more female partners in proportion, those with high income and homosexuality had more fixed male partners in proportion, and those who were married, non-Shanghai domicile had more multiple or occasional male partners in proportion. In the past 6 months, MSM who had unprotected sexual behaviors with female partners, fixed male partners, multiple or occasional male partners accounted for 75.21％, 65.97％, 59.66％, and 48.40％respectively. And analysis showed that respondents had more high-risk sexual behaviors with multiple male partners with their age increasing. [Conclusion] Sexual partner network of MSM is complex. MSM has a variety of sexual partners. Multiple sexual partners and unprotected sexual behaviors exist extensively among MSM. Related factor of sexual behaviors with multiple sexual partners is age.

Analysis of 644 Children's Misusage Medicine Characteristics / 中国药房

OBJECTIVE:To reduce children's mistaking medicine and improve the safety of drug use in children.METHODS:Case number of children's mistaking medicine,the number of poisoning,age distribution and types of mistaken medicine in hospital information system of Beijing Children's Hospital of Capital Medical University during 2014-2016 were counted to analyze data characteristics and put forward relevant preventive strategies.RESULTS & CONCLUSIONS:In 3 years,there were 644 cases of children's mistaking medicine,accounting for 30％ of total number of children poisoning (2 100 cases).A high-risk group concentrated in the 1-4 years old children.The types of mistaking medicine were mostly antihypertensive drugs,antipyretic and analgesic drugs,sedative drugs and antiepileptic drugs.Top 3 medicines were clonazepam (29 cases),ibuprofen (28 cases),aeetaminophen and its compound (12 cases).The main causes of mistaking medicine are mainly associated with hyperactivity,curiosity,imitation and other behavioral characteristics,high availability of drug storage,etc.Types of mistaken medicine are commonly used drugs for the elderly in the family and antipyretic analgesics commonly used in children.It is suggested to prevent the occurrence of children's mistaking medicine from aspects of child nursing person,government and medical institutions by reducing the children accessibility of drugs stored at home,designing drug package difficult to open for children,improving medication compliance of child nursing person,etc.

Comparison of the behavior characteristics in SHR, WKY and SD rat models of attention deficit/hyperactivity disorder / 中国实验动物学报

Objective To compare the differences in behavior characteristics among SHR, WKY and SD rat models of attention deficit/hyperactivity disorder (ADHD), and explore an ideal control model of SHR rats.Methods Using open field test to analyze the rat movement distance, speed, wearing numbers and the number of grooming to evaluate the spontaneous movement in SHR, WKY and SD rats.Using the Morris water maze to test the learning and memory ability among the three rat groups.Results The result of open field test showed that the SHR rats had significantly increased (P< 0.01) total amount of exercise, average speed and wearing numbers than WKY and SD rats.Compared with the WKY rats, SD rats had a significantly higher movement distance (P< 0.01), slightly higher movement speed and wearing number (P< 0.05).In the Morris water maze hidden platform period test, the SD rats had a significantly longer latency than the SHR rats (P< 0.05).SD rats showed longer latency distance on the first, third and fourth days of training, as compared with the SHR rats (P< 0.05 or P< 0.01).Compared with the WKY group, SD rats showed a shorter latency distance in each training time (P< 0.05 or P< 0.01).In the probe trial period, the SD rats showed shorter time and distance ratio to the target quadrant than SHR rats (P< 0.05), while significantly longer than the WKY rats (P< 0.05 or P< 0.01).Conclusions There are significant behavioral differences between SHR and WKY rats, showing certain disadvantages in comparison of the two types of rats.To add SD rats as a control group for SHR rats can improve the comparability of behavior characteristics of SHR rats, and to get more objective evaluation of the behavior characteristics of SHR rats.

Impacts of characteristic personal behaviors on physicians' drug prescription behaviors based on KABP model / 中华医院管理杂志

Within the KABP model as a framework,the authors studied these behaviors at the following three levels,namely knowledge and learning,beliefs and attitude,and behavior.Such aspects as the education background and training,drug expertise,drug infonmation sources,personal experience and habits,attitudes,needs of physicians and patients',socio-demographic characteristics were discussed among others to probe into how physicians' personal behavior affects their prescription behavior.Recommendations raised include education and training,provision of correct drug information,greater publicity,better compensation mechanism of public hospitals,strengthened supervision,establishment of the system of prescription audit,etc.