教学课程资源库 更多>>

关键词： Uncertainty   Training   Measurement uncertainty   Computer science   Labeling   Entropy   Data models   Active learning (AL)   clustering environment   informativeness   label acquisition   representativeness   unsupervised domain adaptation (UDA)  

摘要： Despite the significant progress in unsupervised domain adaptation (UDA), the performance of UDA methods is still far inferior to that of the fully supervised ones. In practical scenarios, it is usually feasible to acquire labels on a small portion of the target data through active learning (AL), which aims to train an effective model with as few queried instances as possible. However, due to the domain shift, the instances selected by existing AL algorithms can be uninformative, redundant, or outlying. To address this issue, we propose a novel approach, namely, clustering environment-aware learning (CEAL), for active domain adaptation (ADA). CEAL selects potentially the most valuable instances under domain shift by exploring the informativeness and representativeness of target samples in a clustering environment-aware manner. Specifically, for the informativeness, we not only leverage the knowledge of individual points but also their nearby neighbors, by measuring the proposed clustering environment aware informativeness score (CEAIS), thus ensuring that the selected samples are highly informative. For the representativeness, we design two schemes called point distance release (PDR) and informativeness score difference exclusion (ISDE) to guarantee the diversity and validity of the selected samples. Furthermore, we fully utilize the large amount of unlabeled data from target domain via pseudo labeling and adopt information maximization to improve the reliability of the target pseudo labels, thereby further improving the performance of the model. The effectiveness of our method is empirically verified on various benchmark datasets against recent state-of-the-art algorithms.

关键词： Rural students   Robotics   Curricular integration   Self-efficacy   Social science   COMPUTATIONAL THINKING   GAME DESIGN   EDUCATION   OUTCOMES  

摘要： Integrating computer science (CS) content into existing STEM curricula emerges as a viable solution to broadening rural students' participation in CS, but rural students have a significant proficiency gap in science and mathematics. By focusing on an English language arts course as the context for CS integration, this exploratory study examined the impact of a robotics-based intervention integrated into a high school language arts class on rural students' CS self-efficacy. A convergent mixed methods design was applied to collect and analyze qualitative and qualitative data separately. Quantitative results confirmed that the integrated robotics-based intervention significantly improved rural students' CS self-efficacy. Qualitative findings provided insights on how embodied learning and mastery experience facilitated by the robotics-based intervention fostered rural students' CS self-efficacy. The findings from the two sources of data were integrated, shedding light on the potential of curricular integration in language arts classes for broadening rural student participation in CS education.

关键词： UDL Redesign Cycle for Online Courses   Universal Design for Learning   Instructional design   Online course design   Higher education  

摘要： This article focuses on the practical implementation of Universal Design for Learning (UDL) in an online computer science course, articulating the collaborative efforts between the instructional designer (the first author) and the faculty member to redesign the course using UDL principles. Specific instances of redesigned learning modules and artifacts in adherence to UDL principles and results and insights gained from the implementation are reported. The article concludes with recommendations for higher education faculty and instructional designers to consider when applying UDL in their online courses.

关键词： Task analysis   Trajectory   Probabilistic logic   Noise measurement   Training data   Noise   Approximation algorithms   Computer science   formal languages   machine learning   programming   representation learning   robot learning   robot programming  

摘要： Imitation Learning (IL) is a promising paradigm for teaching robots to perform novel tasks using demonstrations. Most existing approaches for IL utilize neural networks (NN), however, these methods suffer from several well-known limitations: they 1) require large amounts of training data, 2) are hard to interpret, and 3) are hard to refine and adapt. There is an emerging interest in Programmatic Imitation Learning (PIL), which offers significant promise in addressing the above limitations. In PIL, the learned policy is represented in a programming language, making it amenable to interpretation and adaptation to novel settings. However, state-of-the-art PIL algorithms assume access to action labels and struggle to learn from noisy real-world demonstrations. In this paper, we propose Plunder, a novel PIL algorithm that addresses these shortcomings by synthesizing probabilistic programmatic policies that are particularly well-suited for modeling the uncertainties inherent in real-world demonstrations. Our approach leverages an EM loop to simultaneously infer the missing action labels and the most likely probabilistic policy. We benchmark Plunder against several established IL techniques, and demonstrate its superiority across five challenging imitation learning tasks under noise. Plunder policies outperform the next-best baseline by 19% and 17% in matching the given demonstrations and successfully completing the tasks, respectively.

关键词： Task analysis   Representation learning   Semantics   Self-supervised learning   Training   Mutual information   Computer science   Data mining   graph neural networks   graph representation learning   machine learning  

摘要： Contrastive learning has been successfully applied in unsupervised representation learning. However, the generalization ability of representation learning is limited by the fact that the loss of downstream tasks (e.g., classification) is rarely taken into account while designing contrastive methods. In this article, we propose a new contrastive-based unsupervised graph representation learning (UGRL) framework by 1) maximizing the mutual information (MI) between the semantic information and the structural information of the data and 2) designing three constraints to simultaneously consider the downstream tasks and the representation learning. As a result, our proposed method outputs robust low-dimensional representations. Experimental results on 11 public datasets demonstrate that our proposed method is superior over recent state-of-the-art methods in terms of different downstream tasks. Our code is available at https://***/LarryUESTC/GRLC.

关键词： Data Access   High Performance Computing   Technical Boundaries   Landscape   Scientific Research   Scalable   Data Management   Computer Science   Cloud Computing   Technical Challenges   Scientific Findings   Rapid Advances   Management Costs   Security Measures   Computational Capabilities   Cloud Environment   Foster Collaboration   Cloud Providers   Software Stack   High Performance Computing Systems   Data Security   Unique System   Central Hub   Application Programming Interface   Cloud Platform   Public Cloud   Comprehensive Training   Containerized   Research Community   Pricing Model  

摘要： At the IEEE/ACM International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC23), held in Denver, experts discussed the convergence of high-performance computing and cloud computing. They explored how this integration could address current scientific computing limitations, enhance computational capabilities, and foster global collaboration while focusing on economic, security, technical, and community challenges and opportunities.

关键词： Blockchains   Engines   Synchronization   Semantics   Query processing   Metadata   Computer science   Blockchain system   Ethereum   key-value (KV) store   log-structure merged tree (LSM-tree)   storage engine   INTERNET  

摘要： Blockchain systems play an important role in distributed ledgers, database systems, etc. As more and more blocks are mined, the storage burden of blockchain system is significantly increased. The current blockchain system uniformly transforms all its data into key-value (KV) items and stores them to the underlying log-structure merged tree (LSM-tree) storage engine ignoring the software semantics. Consequently, it not only aggravates the write amplification effect of the storage engine but also increases the redundancy of data query steps, resulting in the performance bottleneck of blockchain system. In this article, we propose a semantic-integrated LSM-tree-based KV storage engine for blockchain systems, called Block-LSM, which significantly improves the data synchronization and data query efficiency of blockchain system. Specifically, we first design a shared prefix scheme to transform blockchain data into ordered KV pairs to alleviate the key range overlaps of different levels in the underlying LSM-tree-based storage engine. Moreover, we propose to maintain several semantic-orientated memory buffers to isolate different kinds of blockchain data and implement memory buffer space management strategy to further improve memory efficiency. To save space overhead, Block-LSM further aggregates multiple blocks into a group and assigns the same prefix to all KV items from the same block group. We also reduce step redundancy in transaction queries by modifying the body data storage format. Finally, we implement Block-LSM in a real blockchain environment and conduct a series of comparative experiments with the typical blockchain system Ethereum. The evaluation results show that Block-LSM significantly reduces up to 7.56x storage write amplification and increases throughput by 8.64x compared with the original Ethereum design. In terms of data lookups (i.e., transaction and account lookup), Block-LSM improves the throughput by 50% compared to the original Ethereum design.

关键词： Computer science and engineering education   Self-directed learning   Self-management   Self-monitoring   Motivation Online learning   PROBLEM-SOLVING SKILLS   ACADEMIC PROCRASTINATION   UNIVERSITY-STUDENTS   COLLEGE-STUDENTS   EDUCATION   PERFORMANCE   TECHNOLOGY   ENGAGEMENT   MOTIVATION   EFFICACY  

摘要： There is an increasing need for computer science and engineering (CSE) online education. This study examined CSE online learners' perceptions of self-directed learning (SDL) readiness, strategies, and satisfaction. The researchers surveyed 225 students and conducted semi-structured interviews with 15 online CSE students. The quantitative survey data and the qualitative interview data were analyzed using descriptive statistics and thematic analysis, respectively. The findings were as follows: (1) Both extrinsic motivation strategies (e.g., future career development, building a learning routine) and intrinsic motivation strategies (e.g., interest in learning) were utilized; (2) Diverse metacognition strategies (e.g., assignments, quizzes, and tests, discussing with peers, tracking progress, staying in touch with professors and teaching assistants) and cognitive strategies (e.g., watching recorded lectures, taking notes, reading books, seeking out information) were used to monitor learning; (3) Time -management (e.g., priority, checklist, fixed schedule, time -block for study) and resource management strategies (e.g., focused on lectures, textbook reading) were leveraged; (4) Specific designs or instruction promoted students' SDL skills, such as access to documented learning materials, reminders sent from instructors, availability of the instructor, group interaction, and flexibility, and (5) Student satisfaction depended on the design of the course. The advantages and disadvantages of online learning were identified. The findings indicated that the instruction strategies and online course design are critical for CSE students' SDL.

关键词： Travel decision making   Recommender systems   Decision support systems   Flickr   Concept modelling   SYSTEMS   SERENDIPITY   ARTIFACT   CULTURE  

摘要： Travel location recommendation systems have long been used by travellers for their ability to suggest destinations and potential travel experiences that match travellers' desires. Recently, a new type of domestic travel has emerged, namely domestic overseas-like travel experiences. These experiences are attractive to travellers who have a preference for exotic locations but no desire to travel internationally. Existing travel recommendation systems were not designed for such applications, nor do they have the relevant ability to recommend domestic overseas-like travel experiences to support travel decision making. To address this challenge, this paper focuses on the development of a recommendation model based on the visual content of photos for domestic overseas-like travel experiences and a prototype application. The application uses the latest advancement in computer vision - the concept model - to learn high-level concepts in an overseas travel destination photo collection to identify similar domestic travel experiences. We demonstrate the usability of the prototype application with a large-scale data set of approximately 479,000 travel photos taken in several countries and evaluate its utility and efficacy through four focus groups with target users.

关键词： Optimization   Statistics   Sociology   Pareto optimization   Computer science   Systematics   History   Archive   archiving methods   environmental selection   evolutionary computation   multiobjective optimization   population update   EVOLUTIONARY ALGORITHM   SUBSET-SELECTION   2-ARCHIVE ALGORITHM   EFFICIENT SOLUTIONS   PARETO   OPTIMIZATION   CONVERGENCE   HYPERVOLUME   INDICATOR   DECOMPOSITION  

摘要： Most multiobjective optimization algorithms maintain an archive explicitly or implicitly during their search. Such an archive can be solely used to store high-quality solutions presented to the decision maker, but in many cases may participate in the search process (e.g., as the population in evolutionary computation). Over the last two decades, archiving, the process of comparing new solutions with previous ones and deciding how to update the archive/population, stands as an important issue in evolutionary multiobjective optimization (EMO). This is evidenced by constant efforts from the community on developing various effective archiving methods, ranging from conventional Pareto-based methods to more recent indicator-based and decomposition-based ones. However, the focus of these efforts is on empirical performance comparison in terms of specific quality indicators; there is lack of systematic study of archiving methods from a general theoretical perspective. In this article, we attempt to conduct a systematic overview of multiobjective archiving, in the hope of paving the way to understand archiving algorithms from a holistic perspective of theory and practice, and more importantly providing a guidance on how to design theoretically desirable and practically useful archiving algorithms. In doing so, we also present that archiving algorithms based on weakly Pareto-compliant indicators (e.g., $\epsilon $ -indicator and IGD+), as long as designed properly, can achieve the same theoretical desirables as archivers based on Pareto-compliant indicators (e.g., hypervolume indicator). Such desirables include the property limit-optimal, the limit form of the possible optimal property that a bounded archiving algorithm can have with respect to the most general form of superiority between solution sets.