Evolutionary computation for addressing high-dimensional expensive problems (HEPs) characterized by both high-dimensional decision variables and resource-intensive evaluations is an important area. In this study, we i...
详细信息
ISBN:
(纸本)9789819771837;9789819771844
Evolutionary computation for addressing high-dimensional expensive problems (HEPs) characterized by both high-dimensional decision variables and resource-intensive evaluations is an important area. In this study, we introduce a novel approach, namely the Hierarchical Diffusion teaching-learning-based Optimizer with Variational autoencoder (HDTOV). Firstly, we employ a variational autoencoder to reduce problem dimensions and facilitate the learning of the optimization process. Secondly, we employ a hierarchical population reconstruction strategy to enhance population diversity. Lastly, to exploit the population more effectively, we implement a diffusion mechanism to prevent premature convergence. The proposed method is validated through experiments on a real-life optimization problem arising from the operation of mobile edge computing systems. The experimental results demonstrate the efficacy and efficiency of HDTOV in addressing HEPs by its outperforming the state of the art.
The financial market is inherently complicated and dynamic, which has increased interest in adapting machine learning (ML) approaches to stock market forecasting. The body of research on machine learning-based stock m...
详细信息
With the world moving forward, leaders in the university sector must proactively innovate in the delivery of teaching programs. The way students learn is changing, such as reduced lecture attendance and a growing dema...
详细信息
ISBN:
(纸本)9781665453318
With the world moving forward, leaders in the university sector must proactively innovate in the delivery of teaching programs. The way students learn is changing, such as reduced lecture attendance and a growing demand for digital learning. This impacts staff workload and can negatively affect both the in-class and online student experience. We need to find ways to enhance student learning and engagement in an attendance-agnostic manner. In this work-in-progress paper, we describe the process of re-designing three courses from the Mathematical Sciences Department at Auckland University of Technology using the flipped classroom and active learning pedagogies. We reflect on the implementation of these courses, which were flipped for one semester (12 weeks) with students who had no previous experience with flipped learning. Despite challenges in involving students in pre-class tasks due to their unfamiliarity, positive feedback on lecture recordings and scaffolding of activities was received from motivated students. Future enhancements of each course will be incorporating students' feedback and the lessons learnt from the course's first run.
Traffic engineering (TE) mechanisms are crucial for achieving optimal levels of network performance over wide-area networks across geographically distributed datacenters. Existing work on traffic engineering formulate...
详细信息
ISBN:
(纸本)9798350386066;9798350386059
Traffic engineering (TE) mechanisms are crucial for achieving optimal levels of network performance over wide-area networks across geographically distributed datacenters. Existing work on traffic engineering formulated the challenges at hand as combinatorial optimization problems, which could take hours to compute on modern wide-area network topologies at the scale of thousands of nodes. To improve the performance of TE mechanisms, we introduce DeepTE, a new TE framework based on machine learning (ML) that is designed for the best possible scalability and performance, capable of completing the computation within milliseconds with networks involving thousands of nodes, and of generating near-optimal TE policies while guaranteeing that all constraints are satisfied. DeepTE is also designed with a distributed ML model architecture, which can be horizontally scaled up to multiple GPUs for even better performance. With real-world traffic matrices, our extensive array of performance evaluations of DeepTE on various network topologies and TE problems show that DeepTE is capable of producing policies within 5% of the optimal results while offering up to 100x performance improvements over state-of-the-art traffic engineering mechanisms.
Natural sciences are important for humanity and our knowledge of the world. A key factor for effective teaching that can lead to knowledge and improvement of learning outcomes is the promotion of students' active ...
详细信息
ISBN:
(数字)9783031533822
ISBN:
(纸本)9783031533815;9783031533822
Natural sciences are important for humanity and our knowledge of the world. A key factor for effective teaching that can lead to knowledge and improvement of learning outcomes is the promotion of students' active participation, the activation of their creativity and their encouragement to implement their own ideas in the educational process, based on their own experiences and individual learning styles. Teachers are suggested to adapt their teaching practices based on their teaching subjects, the profile and learning styles of their students, and the teaching goals they have set for any particular subject. The flipped classroom approach and the creation/visualization of students' mental representations/models in natural sciences were applied as an alternative form of teaching in order to strengthen students' creativity, positive attitude towards science and their cognitive level. Students of a University Pedagogical Department, as being prospective primary education teachers, participated in the research. For the purposes of the study, a mixed method research approach (qualitative and quantitative) was used. The results of the focus group discussion and personal interviews as well as students' final exams revealed a number of interesting findings arising from the application of the specific technique and tools (flipped classroom, self-made models/constructions/visualization of scientific concepts). According to the results, students' creativity was highlighted and enhanced, their active participation in the educational procedure increased significantly and their attitude towards science was positively modified, followed by improved cognitive results.
Natural Language Processing (NLP) has undergone a remarkable transformation, with representation learning playing a pivotal role in reshaping the field. This review explores the evolution of NLP representation learnin...
详细信息
This abstract explores the utilization of deep learning for detecting driver somnolence, aiming to enhance driver safety and alertness monitoring. It investigates the integration of computer vision, physiological sign...
详细信息
The profile of the graduate allows to understand the objective of training during the teaching-learning process in the higher educational institutions, which, in turn, shows the competences acquired by the students. I...
详细信息
ISBN:
(纸本)9798350366716;9798350366709
The profile of the graduate allows to understand the objective of training during the teaching-learning process in the higher educational institutions, which, in turn, shows the competences acquired by the students. In order to predict the achievement of the captured profile, the present research sought to implement a model in machine learning to predict the achievement of the profile of the graduate in the students of the Faculty of Systems engineering of the National University of the Center of Peru (hereinafter nominated as FIS-UNCP), having as input indicators academic performance. The research was carried out following the scientific method, of a quantitative approach, explanatory-predictive level, where the design was nonexperimental longitudinal trend, the chosen population took into account the 355 students of the FIS-UNCP enrolled during the 2023-II academic semester. After the investigation, the result was obtained that the predictive model presents a 97.50% accuracy in the forecasts, with a specificity of 100%, evidencing that the model succeeds in predicting the achievement of the profile of the graduate having as inputs to the indicators of the academic performance of the students. In conclusion, with a T-value=3.239 and P-value=0.001, it can be stated that academic performance significantly influences the achievement of the graduate profile based on a prediction model with machine learning in FIS-UNCP students.
This paper introduces a novel system for multilingual education, leveraging Optical Character Recognition (OCR) and Artificial Intelligence (AI). The system aims to democratize access to educational resources by extra...
详细信息
This paper proposes and implements a feedback system of ideological and political teaching resource base based on emotion analysis and deep learning, aiming to capture students' emotional responses to teaching res...
详细信息
暂无评论