教学课程资源库 更多>>

关键词： Forest fragmentation   Deforestation   Satellite data   Landscape metrics   Landscape fragmentation tool  

摘要： This study investigated the spatial and temporal changes of forest area and fragmentation in Quang Nam province, Vietnam, from 1988 to 2022, using remote sensing and Geographic Information System (GIS) techniques. By analyzing multi-temporal Landsat imagery and applying the Landscape Fragmentation Tool (LFT), we assessed changes in forest cover and the extent of forest fragmentation. The results showed significant changes in forest area, which initially increased due to reforestation programs, then decreased from 2010 to 2022, mainly due to urban expansion, encroachment on agricultural land, and infrastructure development. Landscape metrics, including patch density, edge density, and largest patch index, showed increasing fragmentation, with the area of large core forest shrinking from 80.65% in 1988 to 69.69% in 2022. Fragmented forest classes, such as patch, edge, and perforated forests, increased significantly, indicating rising pressures on forest ecosystems. These changes have negatively affected biodiversity by reducing habitat connectivity and increasing edge impacts. The findings highlight the urgent need for sustainable land use planning, reforestation efforts, and conservation strategies to mitigate further fragmentation and preserve biodiversity. This study has contributed valuable insights into the dynamics of forest change and fragmentation in Vietnam, providing a foundation for future conservation efforts and policy development to maintain ecological integrity in the region.

关键词： ancillary   clustering   computational thinking   curriculum   K-means   random Forest  

摘要： Mathematics courses make up a good percentage of the undergraduate curriculum in the Computer Science and Engineering discipline. It is however important to ascertain how the mathematics courses impact the programming skills of the students. This article examines the impact of mathematics ancillary courses on the Computational Programming Intelligence (CPI) of Computer Science and Engineering students. Using the results of Computer Science students on mathematics and programming courses for seven (7) sessions, Random forest regression and K-means clustering machine learning models were used to study the relationship between the ancillary courses and their performance in programming courses. A Pearson correlation coefficient was computed to assess the linear relationship between five (5) Mathematics ancillary courses and the programming courses. A significant positive correlation (0.29, 0.27, 0.20, 0.10, and 0.09, p = 0.02) was obtained with Linear algebra having the highest and Mathematical methods the least. Consequently, variable importance results show that linear algebra had the highest impact on CPI while Mathematical methods had the least in the following order (29.11%, 20.39%, 19.80%, 17.15%, and 13.55%). Mostly female students of age range 19-20 were found to have been positively impacted more by the mathematics courses. A curriculum guide was presented based on the findings.

关键词： Generative Adversarial Networks   Generators   Training   Gene Expression   Feature Extraction   Data Models   Data Augmentation   Overfitting   Linear Programming   Computer Science   Gene Expression Profile   Generative Adversarial Network   Data Augmentation   Small Sample   Gene Expression Profiles   Data Augmentation   Generative Adversarial Networks   Conditional Generative Adversarial Network   Gradient Penalty   Pearson Correlation   Training Set   Deep Features   Decision Boundary   Nash Equilibrium   Available Gene Expression   Gene Expression Profiling Datasets   Loss Function   Data Distribution   Classification Results   Random Selection   Data Generation   K Nearest Neighbor   Mutual Information   Generative Adversarial Networks Model   Wasserstein Distance   Maximum Mean Discrepancy   Jensen Shannon Divergence   Reproducing Kernel Hilbert Space   Real Data Distribution   Graph Neural Networks   Specific Formula   Information Bottleneck   Discrimination Performance  

摘要： In practical medical diagnosis, small sample sizes in gene expression profiling data can lead to overfitting. Addressing this, we leverage the potential of the Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP) to amplify data volumes. However, it lacks control over the locations of generated samples and struggles to get a better balance between discriminators and generators during training. To overcome these hurdles, we propose the Improved CWGAN-GP, implementing two critical improvements. The first involves the adoption of a data segmentation strategy based on sample influence scores. By calculating the influence score for each sample, we prioritize samples at decision boundaries and outside the distributions as the training set, thus yielding more explicit decision boundaries. The second enhancement is that a depth feature constraint based on the Pearson correlation coefficient is proposed. Here, an encoder extracts the deep features, applying a constraint between the noise and deep features guided by the Pearson correlation coefficient. This strategy navigates the model closer to a Nash equilibrium. Empirical evaluations conducted on six publicly available gene expression profiling datasets validate our approach, demonstrating that it not only generates higher quality samples but also showcases superior stability compared to existing methods.

关键词： Education computing  

关键词： Attention mechanisms   Training   Games   Intelligent transportation systems   Data mining   Computer science   Sports   Knowledge engineering   Human-machine systems   Deep reinforcement learning   Multi-agent reinforcement learning   cooperation   group-specific attention   mixed tasks  

摘要： The cooperation in mixed cooperative-competitive tasks has drawn significant attention in multi-agent deep reinforcement learning. Agents need to cooperate with their teammates while competing against their opponents. However, most existing works treat the cooperative agents and competitive agents equally as they perform the same operation on all the agents. As a result, without distinguishing between cooperative and competitive agents, they may suffer from information disorder in learning an optimally cooperative policy and struggle to decide on the next step action. To address the above issues, we decompose the final Q-value into a weighted combination of three parts: the Q-values of the cooperative agents, the competitive group, and the current agent. A theoretical proof of the correctness of the decomposition is provided. With this decomposition, we are able to consider cooperative and competitive agents separately. Accordingly, we propose a multi-agent actor-critic algorithm called actor-double-attention-critic (ADAC) under centralized training and decentralized execution according to the decomposition. In ADAC, networks with group-specific attention and an attentional weighting network are specially designed. With the designed double-attention structure, ADAC can capture the distributions from different agents and improve cooperation performance. Extensive experiments are conducted in three scenarios with nine settings against six representative methods. The results demonstrate the superiority of the proposed ADAC model against state-of-the-art methods in various mixed cooperative-competitive tasks. The code is available at https://***/CrazyBayes/ADAC

摘要： This essay examines the resulting environmental health conditions in Gaza since October 2023. Evidence is drawn from international agency reports and peer-reviewed literature to document destruction and public health impacts over 18 months. Key themes include water, sanitation, contamination, noise and air pollution, food insecurity, susceptible populations, health system collapse, and trauma. Palestinians are beset with a public health crisis resulting from targeted environmental destruction. Water shortages and sewage contamination increase dehydration, famine, and disease. Air pollution from explosives and waste burning exposes residents to toxins. Waste management has collapsed, heightening disease outbreaks, including polio. Agricultural land is devastated, and food security is threatened with the aid blockade. Power grid destruction has crippled essential services. The health care system's collapse further exacerbates risks, especially for susceptible populations. In addition to mental and physical trauma, cultural and identity loss are immeasurable. The crisis in Gaza has severe and lasting consequences for health, the environment, and regional stability. An end to the bombardment, along with urgent humanitarian aid, release of all hostages, infrastructure rebuilding, and environmental remediation, are essential to mitigate long-term impacts. (AmJ Public Health. 2025;115(7):1053-1061. https://***/10.2105/AJPH.2025.308140)

关键词： Computer science   Human behaviour   Policy  

摘要： Virtual chatbots that simulate conversations with famous actors or sci-fi characters can have real-world consequences.

关键词： Testing   Training   Reinforcement learning   Robot kinematics   Manipulators   Informatics   Computer science   Vectors   Sun   Semantics   Disentangled representation learning   out-of-distribution (OOD) generalization   reinforcement learning (RL)  

摘要： Out-of-distribution (OOD) generalization is critical for applying reinforcement learning algorithms to real-world applications. To address the OOD problem, recent works focus on learning an OOD adaptation policy by capturing the causal factors affecting the environmental dynamics. However, these works recover the causal factors with only an entangled or binary form, resulting in a limited generalization of the policy that requires extra data from the testing environments. To break this limitation, we propose generalizable causal reinforcement learning (GCRL) to learn a disentangled representation of causal factors, on the basis of which we learn a policy that achieves the OOD generalization without extra training. For capturing the causal factors, GCRL deploys a weakly supervised signal with a two-stage constraint to ensure that all factors can be disentangled. Then, to achieve the OOD generalization through causal factors, we establish the dependence of actions on the learned representation and optimize the policy model across multiple environments. Experimental results show that the established dependence recovers the correct relationship between causal factors and actions when the learned policy could address the target tasks in training environments. Benefiting from the recovered relationship, GCRL achieves the OOD generalization on eight benchmarks from Causal World and Mujoco. Moreover, the policy learned by our model is more explainable and can be controlled to generate semantic actions by intervening in the representation of causal factors.

关键词： Critical Thinking   Computer Science   Information Technology Industry   Logical Argument   Artificial Intelligence   Scientific Method  

摘要： Critical thinking-the process of using evidence, observations, and logical arguments to evaluate information, make decisions, or draw conclusions-has long been an important staple of science and engineering. This is particularly true in computing, a field in which emerging technologies increasingly require this skill-both for the producer and the user.

关键词： Convergence   Perturbation methods   Approximation algorithms   Gradient methods   Vectors   Noise measurement   Accuracy   Taylor series   Computer science   Estimation   Balanced generalized simultaneous perturbation stochastic approximation (B-GSPSA)   generalized random directions stochastic approximation (GRDSA)   generalized simultaneous perturbation-based gradient search (GSPGS)   generalized smoothed functional (GSF) procedure   GSPSA   stochastic optimization  

摘要： We present a family of generalized simultaneous perturbation-based gradient search (GSPGS) estimators that use noisy function measurements. The number of function measurements required by each estimator is guided by the desired level of accuracy. We first present in detail unbalanced generalized simultaneous perturbation stochastic approximation estimators and later present the balanced versions of these. We extend this idea further and present the generalized smoothed functional and generalized random directions stochastic approximation estimators, respectively, as well as their balanced variants. We show that estimators within any specified class requiring more number of function measurements result in lower estimator bias. We present a detailed analysis of both the asymptotic and nonasymptotic convergence of the resulting stochastic approximation schemes. We further present a series of experimental results with the various GSPGS estimators on the Rastrigin and quadratic function objectives. Our experiments are seen to validate our theoretical findings.