教学课程资源库 更多>>

关键词： Natural sciences   Applied sciences   Computer science  

摘要： This paper constructs three bi-class multi-relational heterogeneous graphs based on real-world data, and proposes a semantic and relation aware neural network model (SRA-BMHN) designed for bi-class multi-relational heterogeneous graphs. SRA-BMHN consists of two core modules: first, the semantic-aware module, which integrates diverse relational semantic information using a non-linear mapping function and attention mechanism tailored to specific relationship semantics;second, the relation-aware module, which employs a hierarchical bipartite subgraph aggregation strategy. This module first divides the bi-class multi-relational heterogeneous graphs into multiple bipartite subgraphs based on different edge relationship types and captures the topological information within them. It then merges these subgraphs into a relation-weighted sub graph and captures the relational features of nodes on the weighted heterogeneous subgraph. Finally, SRABMHN fuses the features from the semantic-aware module and the relation-aware module to generate the final node embeddings. Experimental results on three datasets demonstrate the superiority of the proposed method.

关键词： Health sciences   Medicine   Computer science   Artificial intelligence applications  

摘要： Deep learning holds great promise in drug discovery, yet its application is hindered by high labeling costs and limited datasets. Developing algorithms that effectively learn from sparsely labeled data is crucial. Capsule networks (CapsNet), introduced in 2017, solve the spatial information loss in traditional neural networks and excel in handling small datasets by capturing spatial hierarchical relationships among features. This capability makes CapsNet particularly promising for drug discovery, where data scarcity is a common challenge. Various modified CapsNet architectures have been successfully applied to drug design and discovery tasks. This review provides a comprehensive analysis of CapsNet's theoretical foundations, its current applications in drug discovery, and its performance in addressing key challenges in the field. Additionally, the study highlights the limitations of CapsNet and outlines potential future research directions to further enhance its utility in drug discovery, offering valuable insights for researchers in both computational and pharmaceutical sciences.

关键词： Students  

摘要： This study explores the integration of Large Language Models (LLMs) into the grading and appeal resolution process in computer science education. We introduce AI-PAT, an AI-powered assessment tool that leverages LLMs to evaluate computer science exams, generate feedback, and address student appeals. AI-PAT was used to assess over 850 exam submissions and handle 185 appeal cases. Our multi-model comparison (ChatGPT, Gemini) reveals strong correlations between model outputs, though significant variability persists depending on configuration and prompt design. Human graders, while internally consistent, showed notable inter-rater disagreement, further highlighting subjectivity in manual evaluation. The appeal process led to grade changes in 74% of cases, indicating the need for continued refinement of AI evaluation strategies. While students appreciated the speed and detail of AI feedback, survey responses revealed trust and fairness concerns. We conclude that AI-PAT offers scalable benefits for formative assessment and feedback, but must be accompanied by transparent grading rubrics, human oversight, and appeal mechanisms to ensure equitable outcomes. Copyright © 2025, The Authors. All rights reserved.

关键词： Physics   Neural networks   Computer science  

摘要： In this study, we explore the inherent trade-off between accuracy and robustness in neural networks, drawing an analogy to the uncertainty principle in quantum mechanics. We propose that neural networks are subject to an uncertainty relation, which manifests as a fundamental limitation in their ability to simultaneously achieve high accuracy and robustness against adversarial attacks. Through mathematical proofs and empirical evidence, we demonstrate that this trade-off is a natural consequence of the sharp boundaries formed between different class concepts during training. Our findings reveal that the complementarity principle, a cornerstone of quantum physics, applies to neural networks, imposing fundamental limits on their capabilities in simultaneous learning of conjugate features. Meanwhile, our work suggests that achieving human- level intelligence through a single-network architecture or massive datasets alone may be inherently limited. Our work provides new insights into the theoretical foundations of neural network vulnerability and opens up avenues for designing more robust neural network architectures.

关键词： Aral Sea   Environmental destruction   Industrial rehabilitation   Livelihood rehabilitation   Forestry and fishery rehabilitation  

摘要： The Aral Sea is a textbook example of an avoidable natural disaster, resulting in the environmental and human devastation of a whole region with wide ranging implications. This study surveyed the people of Kazakhstan on their awareness of the disaster and how they believe it should be addressed. An online survey of 800 people was analyzed using logit regression. It found that about 90% of respondents were aware of the disaster and believed it to be on-going. Over 70% of participants believe it should be restored, along with the livelihood of the impacted communities, even if it involves closing the canals draining the Sea's tributary rivers. Over 60% believed precautionary preemptive measures should be taken in the future to safe guard the environment. These beliefs were strongest among older participants, those with higher levels of education, and those living closer to the region.

关键词： Publishing   Machine learning   Computer science   Physics   Bioinformatics   Software  

摘要： An exclusive Nature analysis reveals the 25 highest-cited papers published this century and explores why they are breaking records. By Helen Pearson, Heidi Ledford, Matthew Hutson and Richard Van Noorden

关键词： Publishing   Machine learning   Computer science   Physics   Bioinformatics   Software  

摘要： Some studies have received hundreds of thousands of citations,Nature’s updated analysis shows.

关键词： Periodic Golay pairs   compression   orderly generation   complementary sequences   algorithms   computer science  

摘要： In this paper, we provide algorithmic methods for conducting exhaustive searches for periodic Golay pairs. Our methods enumerate several lengths beyond the currently known state-of-the-art available searches: we conducted exhaustive searches for periodic Golay pairs of all lengths v <= 72 using our methods, while only lengths v <= 34 had previously been exhaustively enumerated. Our methods are applicable to periodic complementary sequences in general. We utilize sequence compression, a method of sequence generation derived in 2013 by Dokovic and Kotsireas. We also introduce and implement a new method of "multi-level" compression, where sequences are uncompressed in several steps. This method allowed us to exhaustively search all lengths v <= 72 using less than 10 compute years. For cases of complementary sequences where uncompression is not possible, we introduce some new methods of sequence generation inspired by the isomorph-free exhaustive generation algorithm of orderly generation. Finally, we pose a conjecture regarding the structure of periodic Golay pairs and prove it holds in many lengths, including all lengths v < 100. We demonstrate the usefulness of our algorithms by providing the first ever examples of periodic Golay pairs of length v = 90. The smallest length for which the existence of periodic Golay pairs is undecided is now 106.

关键词： STEM education   computer science   education policy   educational equity   causal inference  

摘要： Advanced Placement (AP) provides college-level courses to over 1 million US secondary students annually. Black, Hispanic, and female students have historically been underrepresented in AP Computer Science (CS). A new, broadly focused course-AP CS Principles-launched nationally in 2016-17 with the goal of increasing student participation and diversity. We examine its effects on AP CS participation. Combining publicly available sources, we assemble a panel dataset of annual AP exam-taking and course offerings from 2006-07 to 2020-21 at Massachusetts high schools. Using synthetic difference-in-differences, we estimate that offering the new course led to 16 additional yearly AP Computer Science exams per school, more than tripling baseline exam counts for the average adopting school. Exam counts among female and Black or Hispanic students more than quadrupled. The new exams were concentrated in AP Computer Science Principles, with no statistically significant reduction in exam counts for the preexisting AP CS course. We also estimate that offering the new course increased schools' probability of having any AP CS exam participation by 29 percentage points, with larger gains for female and Black or Hispanic students. We find some evidence of positive spillover effects on several other AP courses. The results suggest the promise of course design and availability in promoting engagement and diversity in advanced STEM education.

关键词： Group theory  

摘要： In the real world, there are many phenomena related to the execution order of events. For example, in a 100 metres, several athletes start at the same starting command. In the downhill skiing event, athletes compete in sequence to decide the winner. Another example is that for a rigid body, if it first moves 10cm forward and then 10cm to the left, or first moves 10cm to the left and then 10cm forward, the total moving distance is 20 cm in both cases. However, when the rigid body has its center of mass fixed in a 3-D coordinate system, if it first rotates 90 degrees in the positive direction of the Y-axis and then 90 degrees in the positive direction of the X-axis, the final posture is different from that when it first rotates 90 degrees in the positive direction of the X-axis and then 90 degrees in the positive direction of the Y-axis. The execution of events serves as the foundation for many disciplines such as physics, computer science, and applied mathematics. Do the athletes in 100 metres start strictly at the same time? Why can we determine the 1st-place finisher in both head-to-head competitions and sequential competitions in speed sports? Why do some events yield the same results when their execution order is changed, while others produce different results? What algebraic properties does the execution order of events have? In this paper, we introduce algebraic theories such as set theory and group theory into the analysis of event execution order. We propose concepts like "optional intervals event" and "sequential operation", summarize their algebraic properties and draw Cayley tables. Then, we use these new concepts to answer the aforementioned questions. Based on these efforts, we offer new interpretations for certain physical phenomena and computer application scenarios. Finally, we present other issues derived from this paradigm. These concepts can deepen our understanding of motion and find applications in areas such as event arrangement, physical simula