Improving Diversity and Quality of Adversarial Examples in Adversarial Transformation Network

This paper proposes a method to mitigate two major issues of Adversarial Transformation Networks (ATN) including the low diversity and the low quality of adversarial examples. In order to deal with the first issue, this research proposes a stacked convolutional autoencoder based on pattern to generalize ATN. This proposed autoencoder could support different patterns such as all-feature pattern , border feature pattern , and class model map pattern . In order to deal with the second issue, this paper presents an algorithm to improve the quality of adversarial examples in terms of L 0 -norm and L 2 -norm. This algorithm employs an adversarial feature ranking heuristics such as JSMA and COI to prioritize adversarial features. To demonstrate the advantages of the proposed method, comprehensive experiments have been conducted on the MNIST dataset and the CIFAR-10 dataset. For the first issue, the proposed autoencoder can generate diverse adversarial examples with the average success rate above 99%. For the second issue, the proposed algorithm could not only improve the quality of adversarial examples significantly but also maintain the average success rate. In terms of L 0 -norm, the proposed algorithm could decrease from hundreds of adversarial features to one adversarial feature. In terms of L 2 -norm, the proposed algorithm could reduce the average distance considerably. These results show that the proposed method is capable of generating high-quality and diverse adversarial examples in practice.

IMPLEMENTASI PEMBELAJARAN BERBASIS HOTS (HIGH ORDER THINKING SKILL) DI SEKOLAH DASAR KOTA MALANG

HOTS (High Order Thinking Skill) is material used by the students for helping the achievement of the competency standard , the purpose of education of primary and competence have (Depdiknas.2006) This study aims to describe the implementation of learning based on High Order Thinking Skills (HOTS) in Elementary Schools in Malang. It is well known that the implementation of the 2013 curriculum suggests that the importance of applying HOTS to the learning process in elementary schools. Therefore, it is necessary to conduct a study in the form of research to see the effectiveness of its implementation in elementary school. This research is a descriptive study with a qualitative approach. There are 2 methods used, namely descriptive and evaluative methods. The study population was all elementary schools in Malang. The research sample there are 40 public elementary schools in all districts in Malang. Data was collected using interview guidelines, observation sheets, and documentation. The results showed: (1) the average level of success of teachers in formulating HOTS-based learning planning in SD Malang only reached 79.46 even though it was in the Good category. (2) The average success rate of the implementation of HOTS-based learning as a whole is at 74.81% with the Fair category. (3) The average success rate of HOTS-based authentic assessment formulation in SD Malang is at 74.65% with the Fair category. Thus, improvement efforts are still needed, especially those related to the implementation of learning and authentic assessment of HOTS-based learning in Malang Primary School.

Application of Intelligent Ultrasound in Real-Time Monitoring of Postoperative Analgesic Nerve Block

In order to monitor the effect of nerve block in postoperative analgesia more accurately, this paper puts forward the application research of ultrasonic real-time intelligent monitoring of nerve block in postoperative analgesia. Ultrasonic real-time intelligent monitoring of nerve block in upper limb surgery, lower limb surgery, and abdominal surgery combined with the nerve stimulator. The experiments show that there are 5 cases of adverse reactions when the nerve stimulator is only used, but no adverse reactions occur when combined with ultrasound-guided block. Continuous subclavian brachial plexus block with the ultrasound-guided nerve stimulator can clearly see the subclavian brachial plexus and its surrounding tissue structure, the direction of needle insertion in the plane, and the diffusion of narcotic drugs. The average success rate of block was up to 95.2%, which was significantly higher than that of nerve stimulator alone, and the success rate of recatheterization after the first failure was also improved. The average postoperative analgesia satisfaction was 85.6%, the average operation time was only 20 min, and the subclavian artery and pleura were avoided effectively. No pneumothorax and other complications occurred. The average success rate of ultrasound-guided subclavicular brachial plexus block in 1-2-year-old children was 97%, which was much higher than the average success rate of nerve stimulator localization with 63%. Ultrasound-guided nerve block not only directly blocks nerves under visual conditions but also helps to observe the structures around nerves and dynamically observe the diffusion of local anesthetics, which can significantly improve the accuracy and success rate of nerve block and reduce the incidence of complications.

Tiny Drone Tracking Framework Using Multiple Trackers and Kalman-based Predictor

Unmanned aerial vehicles like drones are one of the key development technologies with many beneficial applications. As they have made great progress, security and privacy issues are also growing. Drone tacking with a moving camera is one of the important methods to solve these issues. There are various challenges of drone tracking. First, drones move quickly and are usually tiny. Second, images captured by a moving camera have illumination changes. Moreover, the tracking should be performed in real-time for surveillance applications. For fast and accurate drone tracking, this paper proposes a tracking framework utilizing two trackers, a predictor, and a refinement process. One tracker finds a moving target based on motion flow and the other tracker locates the region of interest (ROI) employing histogram features. The predictor estimates the trajectory of the target by using a Kalman filter. The predictor contributes to keeping track of the target even if the trackers fail. Lastly, the refinement process decides the location of the target taking advantage of ROIs from the trackers and the predictor. In experiments on our dataset containing tiny flying drones, the proposed method achieved an average success rate of 1.134 times higher than conventional tracking methods and it performed at an average run-time of 21.08 frames per second.

SECONDARY SCHOOL STUDENTS’ ABILITY TO WORK WITH THE TASKS DESIGNED FOR BIOLOGY OLYMPIAD

For the possibility of using competitive tasks from the Biology Olympiad (BiO), either directly or after certain adaptations for everyday teaching tasks in the teaching process, it was aimed to determine if students could work meaningfully with them. The success of 2nd-year secondary school students (n = 113) in solutions of tasks (n = 5) designated for BiO was compared with the solutions by the regional round BiO´s participants. One-way analysis of variance and Tukey´s multiple comparison test were used for the statistical evaluation of the data. The research shows that secondary school students achieved an average success rate of 47.58 ± 12.51 % and BiO participants 62.69 ± 9.86 %. Within the results of all selected tasks, at least someone of the class of the secondary students achieved similar results in each of the tasks as BiO participants. That indicates that secondary school students were able to work meaningfully with BiO tasks which confirmed students' eligibility for work with these tasks. Keywords: biology education, Biology Olympiad, difficulty of tasks, learning tasks, science education

Optimized Task Group Aggregation-Based Overflow Handling on Fog Computing Environment Using Neural Computing

It is a non-deterministic challenge on a fog computing network to schedule resources or jobs in a manner that increases device efficacy and throughput, diminishes reply period, and maintains the system well-adjusted. Using Machine Learning as a component of neural computing, we developed an improved Task Group Aggregation (TGA) overflow handling system for fog computing environments. As a result of TGA usage in conjunction with an Artificial Neural Network (ANN), we may assess the model’s QoS characteristics to detect an overloaded server and then move the model’s data to virtual machines (VMs). Overloaded and underloaded virtual machines will be balanced according to parameters, such as CPU, memory, and bandwidth to control fog computing overflow concerns with the help of ANN and the machine learning concept. Additionally, the Artificial Bee Colony (ABC) algorithm, which is a neural computing system, is employed as an optimization technique to separate the services and users depending on their individual qualities. The response time and success rate were both enhanced using the newly proposed optimized ANN-based TGA algorithm. Compared to the present work’s minimal reaction time, the total improvement in average success rate is about 3.6189 percent, and Resource Scheduling Efficiency has improved by 3.9832 percent. In terms of virtual machine efficiency for resource scheduling, average success rate, average task completion success rate, and virtual machine response time are improved. The proposed TGA-based overflow handling on a fog computing domain enhances response time compared to the current approaches. Fog computing, for example, demonstrates how artificial intelligence-based systems can be made more efficient.

Why Business Digitalization Is Still So Risky

Due to rapid IT advances and escalated globalization, business digitalization has been accelerating. However, the average success rate of such endevors remains low. As an attempt to reveal why the digital transformation of a business is still so risky, this research analyzes how commonly encountered risks in IS projects are responded to and how the responses affect project outcomes. Resulting from the authors' previous work, they hypothesize that information systems are much more complex as they play an increasingly pivotal role in executing most business functions, and subsequently, the risk pattern of IS projects may have evolved, and that as businesses keep adding more information systems to the enterprise infrastructure, rather than via a piecemeal approach, a framework for enterprise-wide digital integration must be established to guide and evaluate business digitalization. The finding from this research should largely validate their hypotheses and allow them to call for a refocus of the efforts on digitalizing an enterprise.

Design, Development and Testing of Feeding Grippers for Vegetable Plug Transplanters

Vegetable transplanting is an important and advantageous practice in vegetables production systems. In recent years, the development of vegetable transplanting tools has increased, as well as the interest for automatic and robotic transplanters. However, at present, the feeding of transplanting machines is often still performed by hand. This paper presents the design, development and testing of a needle gripper and a two-finger gripper for vegetable transplanting. Both grippers were self-designed and tested for picking, lifting and transplanting plug seedlings. Tests have been conducted on fennel (Foeniculum vulgare L.), leek (Allium ampeloprasum L.) chicory (Cichorium intybus L.) and lettuce (Lactuca sativa L.) seedlings to determine the impact that gripper typology might have on the further growth of plants after transplanting. The average success rate of the two-finger gripper in the transplanting experiment was 95% and of the needle gripper 81.75%, respectively. Although neither gripper typology affected the growth of the seedlings after transplanting, several design implications were identified in order to improve the performance of both grippers. Furthermore, the two-finger gripper is more reliable for lettuce and chicory, while the needle gripper requires root plugs with higher firmness and cohesion to prevent shattering.

1662 Closure of Oro-Antral Fistulae in Conjunction with Functional Endoscopic Sinus Surgery: A Retrospective Analysis + Systematic Review

Aim The aim of this study was to evaluate the surgical closure of OAFs created following dental extractions in conjunction with FESS in adult patients within a hospital in South Wales, UK, during a 5-year period, from 1st January 2014 to 31st December 2018, with a systematic review of the literature to investigate success rates. Method A manual search in the hospital’s theatre system for surgery between 2014 and 2018 containing codes for FESS, OAC or OAF was completed. These were screened for joint cases, and the patient’s hospital numbers entered into a Microsoft Excel spreadsheet. A retrospective analysis of their clinical records was performed. Success was measured as total closure and relief of sinusitis after 1 month. The PRISMA format was used to complete the systematic review. Results 13 patients (mean age 51 ± 12.44, 7:6 male to female) met the inclusion criteria. The most common cause was extraction of the maxillary first molar. A 100% success rate was achieved, with no patients requiring revision surgery. The systematic review highlighted an average success rate of 98.7%. A protocol for the management of OACs was designed in both English and Welsh and distributed to Primary Care Dentists within South Wales. Conclusions OAF closure with FESS can be considered as a highly effective approach, leaving patients symptom free. It is important that protocols are in place for dentists suspecting an OAC in order to expedite patient management.

Object Manipulation with an Anthropomorphic Robotic Hand via Deep Reinforcement Learning with a Synergy Space of Natural Hand Poses

Anthropomorphic robotic hands are designed to attain dexterous movements and flexibility much like human hands. Achieving human-like object manipulation remains a challenge especially due to the control complexity of the anthropomorphic robotic hand with a high degree of freedom. In this work, we propose a deep reinforcement learning (DRL) to train a policy using a synergy space for generating natural grasping and relocation of variously shaped objects using an anthropomorphic robotic hand. A synergy space is created using a continuous normalizing flow network with point clouds of haptic areas, representing natural hand poses obtained from human grasping demonstrations. The DRL policy accesses the synergistic representation and derives natural hand poses through a deep regressor for object grasping and relocation tasks. Our proposed synergy-based DRL achieves an average success rate of 88.38% for the object manipulation tasks, while the standard DRL without synergy space only achieves 50.66%. Qualitative results show the proposed synergy-based DRL policy produces human-like finger placements over the surface of each object including apple, banana, flashlight, camera, lightbulb, and hammer.