Adversarial attacks and robustness in Deep Reinforcement Learning (DRL) have been widely studied in various threat models; however, few consider environmental state perturbations, which are natural in embodied scenarios. To improve the robustness of DRL agents, we formulate the problem of environmental state perturbation, introducing a preliminary non-targeted attack method as a calibration adversary, and then propose a defense framework, named Boosted Adversarial Training (BAT), which first tunes the agents via supervised learning to avoid catastrophic failure and subsequently adversarially trains the agent with reinforcement learning. Extensive experimental results substantiate the vulnerability of mainstream agents under environmental state perturbations and the effectiveness of our proposed attack. The defense results demonstrate that while existing robust reinforcement learning algorithms may not be suitable, our BAT framework can significantly enhance the robustness of agents against environmental state perturbations across various situations.
The technological transition of MacBook charging solutions from MagSafe to USB-C, followed by a return to MagSafe 3, encapsulates the dynamic interplay between technological advancement, environmental considerations, and economic factors. This study delves into the broad implications of these charging technology shifts, particularly focusing on the environmental repercussions associated with electronic waste and the economic impacts felt by both manufacturers and consumers. By investigating the lifecycle of these technologies - from development and market introduction through to their eventual obsolescence - this paper underscores the importance of devising strategies that not only foster technological innovation but also prioritize environmental sustainability and economic feasibility. This comprehensive analysis illuminates the crucial factors influencing the evolution of charging technologies and their wider societal and environmental implications, advocating for a balanced approach that ensures technological progress does not compromise ecological health or economic stability.
Although "climate litigation" is not an indigenous term in China, localizing it is essential to support the development of an independent environmental legal knowledge system in China. Rooted in China's judicial tradition, which emphasizes substantive rationality, traditional legal theories have primarily focused on environmental law. However, the contemporary practices in the rule of law have created an unclear trajectory for climate litigation. Research in this area has long been trapped in a paradigm that relies on lawsuits for ecological environmental damage compensation and environmental public interest litigation, leading to a significant disconnect between theoretical frameworks and practical application. With the advancement of the "dual carbon" strategic goals-carbon peaking and carbon neutrality-it has become imperative to redefine the concept of climate litigation within the Chinese context. We need to establish a theoretical framework that aligns with the "dual carbon" objectives while providing theoretical and institutional support for climate litigation, ultimately contributing to the international discourse on climate justice. Additionally, Hong Kong's proactive climate governance and robust ESG (Environmental, Social, and Governance) practices provide valuable insights for developing comprehensive climate litigation mechanisms. Based on this analysis, we propose concrete plans for building a climate litigation system in China, establishing a preventive relief system and a multi-source legal framework at the substantive level and developing climate judicial mechanisms for mitigation and adaptation at the procedural level.
Government trust, as a core concept in political economy and public policy research, serves as a fundamental cornerstone of democratic legitimacy and state capacity. This paper examines how environmental conditions, particularly sunlight efficiency, influence reported government trust through both affective and cognitive mechanisms. Leveraging World Values Survey Wave 7 data merged with NASA POWER high-frequency weather data, we propose and validate a novel ``salience and attribution'' mechanism: clearer skies may paradoxically reduce government trust by heightening environmental awareness and triggering negative attributions. We further identify potential mediating pathways, including subjective well-being, political interest, political discussion, and health perception, and demonstrate that environmental conditions introduce measurement error in survey-based trust indicators. Our findings provide theoretical contributions to environmental psychology, behavioral political economy, and survey methodology, and yield practical implications for governance, policy design, and survey
Zero-shot learning enables models to generalise to unseen classes by leveraging semantic information, bridging the gap between training and testing sets with non-overlapping classes. While much research has focused on zero-shot learning in computer vision, the application of these methods to environmental audio remains underexplored, with poor performance in existing studies. Generative methods, which have demonstrated success in computer vision, are notably absent from zero-shot environmental sound classification studies. To address this gap, this work investigates generative methods for zero-shot learning in environmental audio. Two successful generative models from computer vision are adapted: a cross-aligned and distribution-aligned variational autoencoder (CADA-VAE) and a leveraging invariant side generative adversarial network (LisGAN). Additionally, we introduced a novel diffusion model conditioned on class auxiliary data. Synthetic embeddings generated by the diffusion model are combined with seen class embeddings to train a classifier. Experiments are conducted on five environmental audio datasets, ESC-50, ARCA23K-FSD, FSC22, UrbanSound8k and TAU Urban Acoustics 2019, and one music classification dataset, GTZAN. Results show that the diffusion model outperforms all baseline methods on average across six audio datasets. This work establishes the diffusion model as a promising approach for zero-shot learning and introduces the first benchmark of generative methods for zero-shot environmental sound classification, providing a foundation for future research.
Lane detection (LD) is an essential component of autonomous driving systems, providing fundamental functionalities like adaptive cruise control and automated lane centering. Existing LD benchmarks primarily focus on evaluating common cases, neglecting the robustness of LD models against environmental illusions such as shadows and tire marks on the road. This research gap poses significant safety challenges since these illusions exist naturally in real-world traffic situations. For the first time, this paper studies the potential threats caused by these environmental illusions to LD and establishes the first comprehensive benchmark LanEvil for evaluating the robustness of LD against this natural corruption. We systematically design 14 prevalent yet critical types of environmental illusions (e.g., shadow, reflection) that cover a wide spectrum of real-world influencing factors in LD tasks. Based on real-world environments, we create 94 realistic and customizable 3D cases using the widely used CARLA simulator, resulting in a dataset comprising 90,292 sampled images. Through extensive experiments, we benchmark the robustness of popular LD methods using LanEvil, revealing substantial performance degradation (-5.37% Accuracy and -10.70% F1-Score on average), with shadow effects posing the greatest risk (-7.39% Accuracy). Additionally, we assess the performance of commercial auto-driving systems OpenPilot and Apollo through collaborative simulations, demonstrating that proposed environmental illusions can lead to incorrect decisions and potential traffic accidents. To defend against environmental illusions, we propose the Attention Area Mixing (AAM) approach using hard examples, which witness significant robustness improvement (+3.76%) under illumination effects. We hope our paper can contribute to advancing more robust auto-driving systems in the future. Website: https://lanevil.github.io/.
Louafi Boutaina, Slimani Chaimae, Bessi Aymane
et al.
Silybum marianum L. Gaertn is a spontaneous plant whose medicinal properties have been used for over two thousand years. This study aims to clarify the understanding and utilisation of S. marianum by the rural and urban populations of Ouezzane region in Morocco, in order to assess the level of recognition and exploitation of this plant. An ethnobotanical survey in this region involved a sample of 140 individuals. Survey results are analyzed using SPSS. The survey results have revealed a significantly limited level of appreciation for S. marianum. Through the use of chi-square statistical tests, we identified significant relationships between our variables and the knowledge about S. marianum and its use. Based on the findings of our study, Silybum marianum L. remains one of Morocco's most neglected and underutilized plants. This may be due to a lack of knowledge or adequate information about its applications, a lack of general interest or even socio-economic factors that limit its exploitation.
Mohammad Erfatpour, Dustin MacLean, Rachid Lahlali
et al.
The ovule is a plant structure that upon fertilization, transforms into a seed. Successful fertilization is required for optimum crop productivity and is strongly affected by environmental conditions including temperature and precipitation. Climate change refers to sustained changes in global or regional climate patterns over an extended period, typically decades to millions of years. These shifts can result from natural processes like volcanic eruptions and solar radiation fluctuations, but in recent times, human activities—especially the burning of fossil fuels, deforestation, and industrial emissions—have accelerated the pace and scale of climate change. Human-induced climate change impacts the agricultural sector mainly through global warming and altering weather patterns, both of which create conditions that challenge agricultural production and food security. With food demand projected to sharply increase by 2050, urgent action is needed to prevent the worst impacts of climate change on food security and allow time for agricultural production systems to adapt and become more resilient. Gaining insights into the female reproductive part of the flower and seed development under extreme environmental conditions is important to oversee plant evolution, agricultural productivity, and food security in the face of climate change. This review summarizes the current knowledge on plant reproductive development and the effects of temperature and water stress, soil salinity, elevated carbon dioxide, and ozone pollution on the female reproductive structure and development across grain legumes, cereal, oilseed, and horticultural crops. It identifies gaps in existing studies for potential future research and suggests suitable mitigation strategies for sustaining crop productivity in a changing climate.
Nutrition. Foods and food supply, Food processing and manufacture
This study examines the fly ash from Soc Son municipal waste power plant (SMPP) and suggests ways to repurpose it to reduce its environmental impact. Fly ash from the Soc Son waste power plant has a gray color, spherical particles with a 5–103 μ m diameter, and a high carbon and heavy metal content. Bermorite crystals can absorb and release heavy metals, making monitoring secondary pollutants during incineration crucial. The EDX analysis of fly ash from the Soc Son waste power plant revealed that it was predominantly contaminated with metal elements, with the highest percentage of calcium. The EDX was able to detect heavy metals in incinerator fly ash. The concentration of Zn in the fly ash exceeded QCVN 07:2009/BTNMT standards, indicating the high amounts of some elements that may be hazardous to the environment and human health. Using the SEM/EDX and XRF, the fly ash from the Soc Son landfill power plant was analyzed and discovered that it exceeds permissible limits for dangerous heavy elements. The most common inorganic elements are Ca, followed by Zn, Pb, Cd, and Ag. Fly ash is classed as hazardous waste due to its high concentration of heavy metals, which results from the combustion of municipal solid waste that has not been separated. Vietnam fights municipal solid waste incinerator fly ash production. Some nations stabilize fly ash to remove harmful components and use it in buildings. Stabilized fly ash makes unfired construction bricks and cement manufacturing components and combining fly ash with inorganic trash protects the environment.
Christina M Post, Jason R Myers, Bethany Winans
et al.
AbstractDevelopmental exposures can influence life-long health; yet, counteracting negative consequences is challenging due to poor understanding of cellular mechanisms. The aryl hydrocarbon receptor (AHR) binds many small molecules, including numerous pollutants. Developmental exposure to the signature environmental AHR ligand 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) significantly dampens adaptive immune responses to influenza A virus in adult offspring. CD8+ cytotoxic T lymphocytes (CTL) are crucial for successful infection resolution, which depends on the number generated and the complexity of their functionality. Prior studies showed developmental AHR activation significantly reduced the number of virus-specific CD8+ T cells, but impact on their functions is less clear. Other studies showed developmental exposure was associated with differences in DNA methylation in CD8+ T cells. Yet, empirical evidence that differences in DNA methylation are causally related to altered CD8+ T-cell function is lacking. The 2 objectives were to ascertain whether developmental AHR activation affects CTL function, and whether differences in methylation contribute to reduced CD8+ T-cell responses to infection. Developmental AHR triggering significantly reduced CTL polyfunctionality, and modified the transcriptional program of CD8+ T cells. S-adenosylmethionine, which increases DNA methylation, but not Zebularine, which diminishes DNA methylation, restored polyfunctionality and boosted the number of virus-specific CD8+ T cells. These findings suggest that diminished methylation, initiated by developmental exposure to an AHR-binding chemical, contributes to durable changes in antiviral CD8+ CTL functions later in life. Thus, deleterious consequence of development exposure to environmental chemicals is not permanently fixed, opening the door for interventional strategies to improve health.
Diana Koldasbayeva, Polina Tregubova, Mikhail Gasanov
et al.
With the rise of electronic data, particularly Earth observation data, data-based geospatial modelling using machine learning (ML) has gained popularity in environmental research. Accurate geospatial predictions are vital for domain research based on ecosystem monitoring and quality assessment and for policy-making and action planning, considering effective management of natural resources. The accuracy and computation speed of ML has generally proved efficient. However, many questions have yet to be addressed to obtain precise and reproducible results suitable for further use in both research and practice. A better understanding of the ML concepts applicable to geospatial problems enhances the development of data science tools providing transparent information crucial for making decisions on global challenges such as biosphere degradation and climate change. This survey reviews common nuances in geospatial modelling, such as imbalanced data, spatial autocorrelation, prediction errors, model generalisation, domain specificity, and uncertainty estimation. We provide an overview of techniques and popular programming tools to overcome or account for the challenges. We also discuss prospects for geospatial Artificial Intelligence in environmental applications.
Albertina Paula Monteiro, Francisco Barbosa, Amélia Silva
et al.
Based on the legitimacy and stakeholders’ theories, this research aims to analyze the environmental information disclosure of Portuguese companies. Specifically, this study aims to explore the environmental information disclosure level, whether the industry (environmentally sensitive) influences the level of ecological matters disclosure, and whether this impacts the companies' performance/profitability. Using the content analysis technique, we developed two indices to assess the level of environmental disclosures in companies' mandatory and voluntary reporting. In addition, for the relationship between variables analysis, we applied the Process Macro of SPSS software. Study results show that (1) there is a positive evolution in the level of environmental disclosure by Lisbon stock exchange listed companies between the years 2015 and 2017, (2) the industry has no significant relationship with profitability; (3) the environmental disclosure acts as a mediator variable in the relationship between industry and profitability. This research presents differences in the tendency of environmental matters disclosure when prepared under an accounting framework or voluntarily and assesses the mediating role of the environmental disclosure index in the relationship between industry and performance.
Shivani Narwal, Rajesh Dhankhar, Savita Kalshan
et al.
Presence of plastics in the surroundings is ubiquitous, as generation of plastics is booming globally and it gets accumulated in oceans leading to deleterious impacts on marine life, public health and the surrounding environment. Owing to its non-degradable nature, plastic particles remain in surroundings for extended periods which automatically facilitate its out spreading. Therefore, there is a need to shift to bio-based plastics, as bio-based green economy hinges on sustainable employment of bioresources for generating a broad spectrum of products, biofuels, chemicals and bioplastics. Typically bioplastics are synthesized from bio-based resources considered to contribute more to sustainable production of plastic as a part of the circular economy. Bioplastics are luring attention and growing as counterfeit material for petroleum-derived plastics owing to their biodegradability. Recently an engrossed interest has been burgeoning in producing drop-in polymers and new-fangled bioplastics by utilizing lignocellulosic feedstock. This paper reviews the enormous potential of lignocellulosic feedstock as a significant inedible substrate for bioplastic synthesis. Polyhydroxyalkanoates, polyurethanes, polylactic acid and starch-bioplastic are prevailing bio-based plastic comparably derived from lignocellulosic biomass. In forthcoming years bioplastic derived years’ bioplastic derived from lignocellulose will loom as valuable material in numerous fields for an extensive range of cutting-edge applications.
Ayrton Bangolo, Pierre Fwelo, Sowmya Sagireddy
et al.
Background: Primary malignant melanomas of the Gastrointestinal mucosa are uncommon. Most cases of gastrointestinal (GI) melanomas are secondary, arising from metastasis at distant sites. The purpose of this study is to assess to what extent the interaction between independent prognostic factors (age and tumor site) of primary GI melanoma influence survival. Furthermore, we also aimed to investigate the clinical characteristics, survival outcomes, and independent prognostic factors of patients with primary GI melanoma in the past decade. Methods: A total of 399 patients diagnosed with primary GI melanoma, between 2008 and 2017, were enrolled in our study by retrieving data from the Surveillance, Epidemiology, and End Results (SEER) database. We analyzed demographics, clinical characteristics, and overall mortality (OM) as well as cancer-specific mortality (CSM) of primary GI melanoma. Variables with a <i>p</i> value < 0.1 in the univariate Cox regression were incorporated into the multivariate Cox model (model 1) to determine the independent prognostic factors, with a hazard ratio (HR) of greater than 1 representing adverse prognostic factors. Furthermore, we analyzed the effect of the interaction between age and primary location on mortality (model 2). Results: Multivariate cox proportional hazard regression analyses revealed higher OM in age group 80+ (HR = 5.653, 95% CI 2.212–14.445, <i>p</i> = 0), stomach location of the tumor (HR = 2.821, 95% CI 1.265–6.292, <i>p</i> = 0.011), regional lymph node involvement only (HR = 1.664, 95% CI 1.051–2.635, <i>p</i> < 0.05), regional involvement by both direct extension and lymph node involvement (HR = 1.755, 95% CI 1.047–2.943, <i>p</i> < 0.05) and distant metastases (HR = 4.491, 95% CI 3.115–6.476, <i>p</i> = 0), whereas the lowest OM was observed in patients with small intestine melanoma (HR = 0.383, 95% CI 0.173–0.846, <i>p</i> < 0.05). Multivariate cox proportional hazard regression analyses of CSM also revealed higher mortality of the same groups and lower CSM in small intestine and colon melanoma excluding the rectum. For model 2, considering the interaction between age and primary site on mortality, higher OM was found in age group 80+, followed by age group 40–59 then age group 60–79, regional lymph node involvement only, regional involvement by both direct extension and lymph node involvement and distant metastases. The small intestine had a lower OM. The rectum as primary location and the age range 40–59 interacted to lower the OM (HR = 0.14, 95% CI 0.02–0.89, <i>p</i> = 0.038). Age and primary gastric location did not interact to affect the OM. For the CSM, taking into account the interaction between age and the primary location, higher mortality was found in the same groups and the colon location. The primary colon location also interacted with the age group 40–59 to increase the CSM (HR = 1.38 × 10<sup>9</sup>, 95% CI 7.80 × 10<sup>7</sup>–2.45 × 10<sup>10</sup>, <i>p</i> = 0). Conclusions: In this United States population-based retrospective cohort study using the SEER database, we found that only the age range 40–59 interacted with the rectum and colon to lower and increase mortality respectively. Primary gastric location, which was the single most important location to affect mortality, did not interact with any age range to influence mortality. With those results, we hope to shed some light on this rare pathology with a very dismal prognosis.
I teach Phy-Chem and AP Biology in an urban magnet high school. Phy-Chem is understood to be a freshmen general science class with an emphasis on physics and chemistry. AP Biology is a high school biology class that uses a college curriculum provided by the College Board. At my school AP Biology’s, like all AP classes, enrollment is an open-door policy wherein students can choose to take the class and do not have to meet any requirements to enter. The Phy-Chem curriculum is structured around 21 Next Generation Science Standards (NGSS). These standards are drawn from both the physical and environmental science standards as well as engineering standards.
Riccardo Gianluigi Serio, Maria Michela Dickson, Diego Giuliani
et al.
The transition to more environmentally sustainable production processes and managerial practices is an increasingly important topic. Many industries need to undergo radical change to meet environmental sustainability requirements; the tourism industry is no exception. In this respect, a particular aspect that needs further attention is the relationship between airport performances and investments in environmental sustainability policies. This work represents a first attempt to provide empirical evidences about this relationship. Through the application of a non-parametrical method, we first assess the efficiency of the Italian airports industry. Secondly, we investigated the relationship between airports performance and management commitment toward the ecological transition using a Tobit regression model. The results show that airports adherence to formal multi-year ecological transition programs has a positive and consistent impact on their performance.
Potential societal and environmental effects such as the rapidly increasing resource use and the associated environmental impact, reproducibility issues, and exclusivity, the privatization of ML research leading to a public research brain-drain, a narrowing of the research effort caused by a focus on deep learning, and the introduction of biases through a lack of sociodemographic diversity in data and personnel caused by recent developments in machine learning are a current topic of discussion and scientific publications. However, these discussions and publications focus mainly on computer science-adjacent fields, including computer vision and natural language processing or basic ML research. Using bibliometric analysis of the complete and full-text analysis of the open-access literature, we show that the same observations can be made for applied machine learning in chemistry and biology. These developments can potentially affect basic and applied research, such as drug discovery and development, beyond the known issue of biased data sets.