Task Learning Strategies Research Articles

AIR-Net: A novel multi-task learning method with auxiliary image reconstruction for predicting EGFR mutation status on CT images of NSCLC patients

Automated and accurate EGFR mutation status prediction using computed tomography (CT) imagery is of great value for tailoring optimal treatments to non-small cell lung cancer (NSCLC) patients. However, existing deep learning based methods usually adopt a single task learning strategy to design and train EGFR mutation status prediction models with limited training data, which may be insufficient to learn distinguishable representations for promoting prediction performance. In this paper, a novel multi-task learning method named AIR-Net is proposed to precisely predict EGFR mutation status on CT images. First, an auxiliary image reconstruction task is effectively integrated with EGFR mutation status prediction, aiming at providing extra supervision at the training phase. Particularly, we adequately employ multi-level information in a shared encoder to generate more comprehensive representations of tumors. Second, a powerful feature consistency loss is further introduced to constrain semantic consistency of original and reconstructed images, which contributes to enhanced image reconstruction and offers more effective regularization to AIR-Net during training. Performance analysis of AIR-Net indicates that auxiliary image reconstruction plays an essential role in identifying EGFR mutation status. Furthermore, extensive experimental results demonstrate that our method achieves favorable performance against other competitive prediction methods. All the results executed in this study suggest that the effectiveness and superiority of AIR-Net in precisely predicting EGFR mutation status of NSCLC.

A Task-Learning Strategy for Robotic Assembly Tasks from Human Demonstrations.

In manufacturing, traditional task pre-programming methods limit the efficiency of human–robot skill transfer. This paper proposes a novel task-learning strategy, enabling robots to learn skills from human demonstrations flexibly and generalize skills under new task situations. Specifically, we establish a markerless vision capture system to acquire continuous human hand movements and develop a threshold-based heuristic segmentation algorithm to segment the complete movements into different movement primitives (MPs) which encode human hand movements with task-oriented models. For movement primitive learning, we adopt a Gaussian mixture model and Gaussian mixture regression (GMM-GMR) to extract the optimal trajectory encapsulating sufficient human features and utilize dynamical movement primitives (DMPs) to learn for trajectory generalization. In addition, we propose an improved visuo-spatial skill learning (VSL) algorithm to learn goal configurations concerning spatial relationships between task-relevant objects. Only one multioperation demonstration is required for learning, and robots can generalize goal configurations under new task situations following the task execution order from demonstration. A series of peg-in-hole experiments demonstrate that the proposed task-learning strategy can obtain exact pick-and-place points and generate smooth human-like trajectories, verifying the effectiveness of the proposed strategy.

The Role of Aptitude Treatment Interaction Instructional Model with Task and Forced Instructional Strategy on Student Mathematical Reasoning Ability

This research was conducted to determine the effect of aptitude-treatment interaction (ATI) instructional models with task and forced instructional strategy to mathematical reasoning abilities. Mathematical reasoning ability is an ability that belongs to the goals of national education, but its achievement is still fairly low. The method used in this study is a quasi-experimental method with a posttest-only research design. Sampling in this study using purposive sampling technique. The study sample consisted of 60 students divided into two classes (30 experimental classes and 30 control classes). This research was conducted in the school of SMK Perintis 1 Depok and SMK Fajar Depok. The experimental class uses aptitude-treatment interaction (ATI) learning model with task and forced learning strategies, while the control class uses conventional methods. The collected data is then analyzed using the calculation of chi-square test, Fisher test, and hypothesis testing. The results of the data analysis showed that the average value of students who learned using aptitude-treatment interaction (ATI) learning model with task and forced learning strategies was higher than the average value of students who learned using conventional methods. Based on the results of hypothesis testing shows that there is an influence of aptitude-treatment interaction (ATI) learning model with task learning strategies and forced on students' mathematical reasoning abilities.

MODIFIKASI MODEL PEMBELAJARAN THINK TALK WRITE (TTW) DENGAN STRATEGI PEMBELAJARAN TUGAS DAN PAKSA

The learning model that is still teacher-centered causes low activity and student learning outcomes in the classroom, many students do not understand the material taught in the classroom especially if the teacher does not provide additional assignments. This writing aims to improve the activities and student learning outcomes by analyzing the combination of Think Talk Write (TTW) learning models with forced and task learning strategies. To provide a motivation for teachers to package learning that is creative and innovative, making learning activities interesting and not monotonous. The results of this modification indicate that learning with the Think Talk Write (TTW) learning model through forced and task strategies can improve the learning process so that students become more disciplined and do not waste time doing tasks. The advantages of modification of the Think Talk Write (TTW) learning model with forced and task learning strategies is to increase student learning motivation, improve the quality of learning, train students' understanding by giving assignments continuously, teach discipline to students in order to be accountable for tasks assigned, and reduce laziness in students. It is recommended to modify the Think Talk Write (TTW) learning model with forced and forced learning strategies both applied in the classroom so that the application becomes more effective while still taking into account the conditions of the school environment, students and teachers

MODIFIKASI MODEL PEMBELAJARAN MATEMATIKA COLLABORATIVE LEARNING DENGAN STRATEGI TUGAS DAN PAKSA

The purpose of this writing is to make students more active, discipline able to work together, in the learning process and also can improve learning creativity. The learning process in Indonesia is currently too monotonous and runs as it is without focusing on learning. So, this problem must be taken seriously, especially by the teacher. Therefore, the teacher must be able to inspire students to be actively involved in the learning process. The selection and determination of learning strategies that are most appropriate to the characteristics of students will certainly produce the desired learning goals. To make it happen, a merger of Collaborative Learning Mathematical Learning Model with Task and Forced Learning Strategy is combined. Collaborative Learning is a Learning Model that forms groups that makes it easier for teachers and students to complete learning material. Task and Forced Learning Strategies are strategies that focus on giving assignments and a little coercion so that students complete their tasks on time so that the learning process can run effectively. By modifying the Collaborative Learning Mathematics Learning Model with Task Learning Strategies and Forces students will motivate students that learning is the responsibility, all students will be more active, creative, and disciplined in learning, making students not delay work or assignments given, and will make group work becomes solid. To apply the Collaborative Learning Mathematics Learning Model with Task and Forced Learning Strategies teachers should have better and more mature planning. The teacher should master and understand the rules in the Collaborative Learning Mathematical Learning Model with Task and Forced Learning Strategies, and the teacher must always motivate students to be more active in learning.

MODIFIKASI MODEL PEMBELAJARAN MIND MAPPING DENGAN STRATEGI PEMBELAJARAN TUGAS DAN PAKSA

This writing aims to improve students' understanding by modifying learning models with appropriate learning strategies that are considered based on the characteristics and intellectual development of students to improve learning outcomes. The learning process in Indonesia is currently still centered on teachers which causes students' memory and understanding to be still low. This is the background for the purpose of mind mapping learning models, namely making patterned visual and graphical subject matter that can help strengthen and recall information that has been studied. The selection of the right learning strategy can improve the results that will be obtained from the application of learning models in the classroom. Forced and forced learning strategies are chosen to complement and perfect the implementation of mind mapping learning models in the classroom. This strategy aims to train students' independence and discipline in learning through assignments given with clear time limits and strict penalties if there are students not completing their assignments properly. The combination of mind mapping learning models with forced and task learning strategies can be an alternative to improve the quality of learning in schools. With the increase in thinking power accompanied by student discipline in learning, the learning process in the classroom will run well and get maximum results so that the objectives of the learning can be achieved. For educators it is recommended to implement mind mapping learning models with forced and forced learning strategies in schools, so that learning objectives can be achieved optimally.

MODIFIKASI MODEL PEMBELAJARAN QUANTUM LEARNING DENGAN STRATEGI PEMBELAJARAN TUGAS DAN PAKSA

This writing aims to make students more active and disciplined in the learning process and can also increase creativity and learning outcomes. The low mathematics learning outcomes are not only due to difficult mathematics, but are caused by several factors which include students themselves, teachers, learning approaches, and learning environments that are interconnected with each other. To improve the ability and results of learning it is necessary to make modifications to the task learning strategy and force. Quantum learning is a tip, a guide, a strategy and an entire learning process that can sharpen understanding and memory, and make learning a pleasant and useful process. Task and Forced Learning Strategies are strategies that focus on giving assignments and a little coercion so that students complete their tasks on time so that the learning process can run effectively. Therefore, the writer modifies the model of quantum learning with task and forced learning strategies, the results of this modification show that learning with quantum learning models with forced and task strategies can improve the learning process so that students become more disciplined in doing tasks, can motivate student learning, and can improve student learning outcomes.

Effect of Guided Vs. Discovery Learning Strategies on Initial Motor Task Learning, Transfer, and Retention

Abstract The purpose of this study was to compare the effects of guided vs. discovery strategies on (a) the learning of a primary psychomotor task; (b) transfer psychomotor tasks with two different degrees of difficulty; (c) transfer of learning as influenced by guided vs. discovery learning; and (d) retention (relearning) of the primary task as a function of primary and transfer task learning strategies as well as the complexity of the transfer task. Female college students (N = 128) were assigned to either a guided or discovery condition for the learning of the primary task and then to one of four conditions for transfer learning—discovery or guided, for a simple or complex transfer task—to learn computer managed serial motor tasks. Guided learning was most efficient in initial task learning. Efficiency in transfer learning was greatest when the strategy for transfer learning was the same as that used in primary task learning. Neither learning strategy was more effective when considering a transfer task...