by jsendak | Apr 25, 2025 | Art
Inaugurating Art and Freedom: Ai Weiwei’s Monumental Public Art Installation
Ai Weiwei, a renowned artist and activist, is set to establish the inaugural Art X Freedom commission, creating a groundbreaking public art installation that will ignite conversations around the world. This unique project stands as a testament to the enduring power of art to question, challenge, and evoke transformations in society.
Exploring the Intersection of Art and Activism
Since time immemorial, artists have played a pivotal role in shaping the social, cultural, and political landscapes of their times. Thinkers like Leonardo da Vinci, Frida Kahlo, and Pablo Picasso used their creative genius to challenge the status quo, unravel injustices, and inspire meaningful change.
Today, Ai Weiwei carries this torch of artistic activism, fearlessly illuminating the dark corners of authoritarianism, censorship, and human rights abuses. Inspired by his own experiences as a dissident in China, Weiwei’s works span across various media, transcending boundaries between art, architecture, and social commentary.
“The role of the artist is to ask questions, not to answer them.”
Ai Weiwei
A Journey Through Weiwei’s Artistic Endeavors
Weiwei’s oeuvre has left an indelible mark on the contemporary art scene. From the iconic “Sunflower Seeds” installation, where millions of porcelain seeds laid bare the perils of conformity, to “Grass Mud Horse,” a whimsical confrontation against censorship, his creations communicate powerful messages through their sheer scale and symbolism.
Beyond the confines of galleries and museums, Ai Weiwei has embraced the global stage, culminating in the forthcoming Art X Freedom commission. This landmark project marks a dynamic collaboration between art and the public sphere, fostering dialogue, empathy, and ultimately, freedom.
Empowering the Public: The Art X Freedom Commission
The Art X Freedom commission envisions the transformation of a bustling downtown square into a mesmerizing playground of imagination, resilience, and dissent. Through monumental sculptures, immersive installations, and thought-provoking graffiti, Weiwei endeavors to remind us of the power we each possess to shape a more just and equitable world.
By juxtaposing historical narratives with contemporary concerns, Weiwei invites viewers to analyze the triumphs and failures of societies past and present. The Art X Freedom commission seeks to bridge the gap between art and an unsuspecting public, provoking introspection and reinvigorating our collective pursuit of freedom.
“Creativity is part of the human condition. It must always be guarded and cherished, even if society questions its worth.”
Ai Weiwei
With the inauguration of the Art X Freedom commission, Ai Weiwei propels us towards a new era of artistic expression, uncompromising activism, and societal change. In the face of adversity and oppression, this monumental public art installation invites us to unite under the banner of freedom, knowing that art has the power to spark the flames of revolution and pave the way for a better tomorrow.
The inaugural Art X Freedom commission will be a monumental new public art installation by artist and activist Ai Weiwei.
Read the original article
by jsendak | Apr 24, 2025 | AI
arXiv:2504.16635v1 Announce Type: new Abstract: In an environment of increasingly volatile financial markets, the accurate estimation of risk remains a major challenge. Traditional econometric models, such as GARCH and its variants, are based on assumptions that are often too rigid to adapt to the complexity of the current market dynamics. To overcome these limitations, we propose a hybrid framework for Value-at-Risk (VaR) estimation, combining GARCH volatility models with deep reinforcement learning. Our approach incorporates directional market forecasting using the Double Deep Q-Network (DDQN) model, treating the task as an imbalanced classification problem. This architecture enables the dynamic adjustment of risk-level forecasts according to market conditions. Empirical validation on daily Eurostoxx 50 data covering periods of crisis and high volatility shows a significant improvement in the accuracy of VaR estimates, as well as a reduction in the number of breaches and also in capital requirements, while respecting regulatory risk thresholds. The ability of the model to adjust risk levels in real time reinforces its relevance to modern and proactive risk management.
The article “arXiv:2504.16635v1” addresses the challenge of accurately estimating risk in today’s volatile financial markets. Traditional econometric models, such as GARCH, struggle to adapt to the complexity of current market dynamics. To overcome these limitations, the authors propose a hybrid framework for Value-at-Risk (VaR) estimation that combines GARCH volatility models with deep reinforcement learning. By incorporating directional market forecasting using the Double Deep Q-Network (DDQN) model, the authors create an architecture that allows for dynamic adjustment of risk-level forecasts based on market conditions. Empirical validation on daily Eurostoxx 50 data demonstrates significant improvements in the accuracy of VaR estimates, a reduction in breaches, and lower capital requirements while still adhering to regulatory risk thresholds. This model’s ability to adjust risk levels in real-time highlights its relevance to modern and proactive risk management.
Reimagining Risk Estimation: A Hybrid Framework for Value-at-Risk
In today’s ever-changing financial landscape, accurately estimating risk has become a daunting challenge. Traditional econometric models, such as GARCH and its variants, have proven to be insufficient in adapting to the complexity and volatility of the current market dynamics. To overcome these limitations, a hybrid framework for Value-at-Risk (VaR) estimation that combines GARCH volatility models with deep reinforcement learning is proposed. This innovative approach incorporates directional market forecasting using the Double Deep Q-Network (DDQN) model, treating the task as an imbalanced classification problem.
One of the major limitations of traditional econometric models is their reliance on rigid assumptions that do not adequately capture the intricacies of market behavior. The proposed hybrid framework addresses this drawback by leveraging the power of deep reinforcement learning, which enables the dynamic adjustment of risk-level forecasts according to prevailing market conditions.
The architecture of the hybrid framework allows for real-time adjustment of risk levels, offering a proactive approach to risk management that is essential in today’s fast-paced financial markets. By combining GARCH volatility models with deep reinforcement learning, the proposed framework enhances the accuracy of VaR estimates and reduces the number of breaches, as well as the capital requirements, while still adhering to regulatory risk thresholds.
Empirical validation of the hybrid framework using daily Eurostoxx 50 data, encompassing periods of crisis and high volatility, demonstrated a significant improvement in the accuracy of VaR estimates. This finding highlights the potential of the hybrid framework to better capture market dynamics and provide more reliable risk estimations.
The ability of the hybrid framework to adapt to changing market conditions and adjust risk levels in real time is a game-changer in the field of risk management. Traditional models often fail to account for shifts in market dynamics, resulting in inaccurate risk estimations that may lead to substantial losses. The integration of deep reinforcement learning into the risk estimation process offers a more robust and flexible approach that better aligns with the complexities of today’s financial markets.
As financial markets continue to evolve, embracing innovative solutions becomes imperative for effective risk management. The proposed hybrid framework for VaR estimation, combining GARCH volatility models with deep reinforcement learning, offers a forward-thinking approach that can enhance risk management practices. By leveraging the power of artificial intelligence and machine learning, financial institutions can achieve more accurate risk estimations, reduce breaches, and ensure compliance with regulatory requirements.
In conclusion, the hybrid framework presented in this article provides a fresh perspective on risk estimation in volatile financial markets. By incorporating deep reinforcement learning with GARCH volatility models, the proposed framework enables dynamic adjustment of risk-level forecasts and offers real-time risk management capabilities. This innovative solution holds great promise for improving the accuracy of VaR estimates and strengthening risk management practices in the face of evolving market dynamics.
The paper titled “A Hybrid Framework for Value-at-Risk Estimation using GARCH and Deep Reinforcement Learning” addresses the challenge of accurately estimating risk in volatile financial markets. The authors argue that traditional econometric models like GARCH are often too rigid to adapt to the complexity of current market dynamics. To overcome these limitations, they propose a hybrid framework that combines GARCH volatility models with deep reinforcement learning.
The incorporation of deep reinforcement learning into the estimation of Value-at-Risk (VaR) is an interesting approach. By using the Double Deep Q-Network (DDQN) model, the authors aim to incorporate directional market forecasting into the framework. They treat the task as an imbalanced classification problem, which allows for dynamic adjustment of risk-level forecasts based on market conditions.
The empirical validation of the proposed framework using daily Eurostoxx 50 data covering periods of crisis and high volatility is a significant contribution. The results show a significant improvement in the accuracy of VaR estimates, as well as a reduction in the number of breaches and capital requirements, while still respecting regulatory risk thresholds.
One of the key strengths of this hybrid framework is its ability to adjust risk levels in real-time. This is particularly relevant in modern risk management practices, where proactive risk mitigation is crucial. By incorporating deep reinforcement learning, the model can adapt to changing market dynamics and provide more accurate risk estimates.
However, it is important to note that the paper does not discuss potential limitations or challenges of implementing this hybrid framework in real-world scenarios. It would be valuable to explore how the model performs in different market conditions and whether it can be effectively used by financial institutions for risk management purposes.
Overall, the proposed hybrid framework for VaR estimation shows promising results in improving accuracy and reducing breaches and capital requirements. It provides a novel approach to incorporating machine learning techniques into risk management practices. Future research can focus on further validating the framework with different datasets and exploring its practical implementation in financial institutions.
Read the original article
by jsendak | Apr 19, 2025 | Science
Published on 17 April 2025, a recent Nature article titled “Freshwater insects used ‘microplastic’ as a building material long before scientists coined the term” sheds light on an intriguing phenomenon. It highlights how insects have been using microplastics as a building material long before scientists even recognized the existence of such particles.
Key Points:
- Freshwater insects were observed incorporating microplastics into their structures before the term ‘microplastics’ was coined.
- Insects have been utilizing microplastics as a building material for their nests and dwellings for an extended period.
- This discovery raises concerns about the potential impacts of microplastics on insect populations and ecosystems.
- Further research is needed to understand the extent of microplastic incorporation by insects and its ecological implications.
Potential Future Trends:
This intriguing discovery opens up a realm of potential future trends related to microplastics and their interaction with insects and freshwater ecosystems. Here are some of the possible developments we may witness:
1. Increased Awareness and Research:
The recognition of insects using microplastics as a building material will likely increase awareness and stimulate further research in this field. Scientists may delve deeper into the mechanisms underlying this behavior and study the potential consequences for insect populations and ecosystems.
2. Impact on Insect Behavior and Ecology:
Understanding the extent to which insects incorporate microplastics into their structures can provide valuable insights into their behavior and ecology. It may help researchers determine whether microplastics affect vital aspects of their lives, such as reproduction, survival, and population dynamics.
3. Conservation Strategies:
This discovery might compel conservationists to implement novel strategies to mitigate the impact of microplastics on insect populations and freshwater ecosystems. Such initiatives could focus on reducing the release of microplastics into the environment and restoring habitats to promote the well-being of insect species.
4. Sustainable Material Alternatives:
The recognition of insects utilizing microplastics as a building material could fuel the development and implementation of sustainable alternatives. Scientists and engineers may strive to create eco-friendly materials that mimic the characteristics of microplastics to offer insects suitable alternatives for their construction needs.
Predictions:
Based on the current understanding and potential future trends, we can make the following predictions:
1. Heightened Research Efforts:
As scientists become increasingly aware of insects’ use of microplastics, we can anticipate a surge in research efforts focusing on understanding the ecological implications of this behavior. This research will likely aim to provide a comprehensive understanding of the long-term effects on insect populations and ecosystems.
2. Innovative Environmental Conservation Approaches:
This discovery will likely lead to the development of innovative environmental conservation approaches regarding microplastics and their impact on freshwater ecosystems. Conservation efforts might focus on reducing microplastic pollution, implementing habitat restoration measures, and identifying strategies to support insect populations affected by microplastics.
3. Integration of Sustainable Materials:
With the recognition of insects utilizing microplastics, there may be a significant push towards the integration of sustainable material alternatives in various industries. This could lead to eco-friendly products and materials that limit the release of microplastics into the environment while still meeting the needs of society.
Recommendations for the Industry:
Considering the potential future trends and the importance of addressing the impact of microplastics on freshwater ecosystems, here are some recommendations for the industry:
- Research and Development: Industries must invest in research and development initiatives focused on eco-friendly materials and efficient waste management. Collaboration with academia and environmental organizations can help drive innovation in sustainable practices.
- Public Awareness: Raising public awareness about the detrimental effects of microplastics and the importance of reducing their use is crucial. Industries should actively participate in educational campaigns and promote responsible consumption and disposal of plastics.
- Regulatory Measures: Governments and regulatory bodies must implement stricter regulations and guidelines regarding the use and disposal of microplastics. Such measures can help limit the release of microplastics into the environment and encourage the adoption of sustainable alternatives.
- Collaboration: Collaboration between industries, research institutions, and environmental organizations is essential to tackle the issue of microplastics effectively. Joint efforts can lead to innovative solutions and strategies that minimize environmental harm.
Conclusion:
The discovery of insects utilizing microplastics as a building material suggests a fascinating interplay between nature and the plastic pollution crisis. This finding has the potential to drive increased research efforts, innovative conservation strategies, sustainable material alternatives, and stricter regulations. By taking proactive measures and embracing sustainable practices, industries can contribute to mitigating the impact of microplastics on freshwater ecosystems, insect populations, and the overall well-being of our planet.
Reference:
Nature, Published online: 17 April 2025; doi:10.1038/d41586-025-01265-y
by jsendak | Apr 18, 2025 | Art
Thematic Preface: Reviving Architectural Heritage – The Armenian Pavilion at the 24th Triennale Milano International Exhibition
In partnership with the National Library of Armenia, the Library for Architecture (LFA) is proud to announce the forthcoming inauguration of the Armenian Pavilion at the prestigious 24th Triennale Milano International Exhibition, taking place on May 13, 2025. This momentous event will serve as a platform to showcase and celebrate the rich architectural heritage of Armenia, tracing its roots back to ancient times while also looking towards the future.
Armenia, a land steeped in history, has a profound architectural legacy that reflects the cultural, social, and political aspects of its diverse past. Nestled at the crossroads of Europe and Asia, the country has endured centuries of conquests, invasions, and cultural assimilation, resulting in a unique amalgamation of styles and influences in its architectural identity.
Throughout history, Armenia has witnessed the rise and fall of empires, from the Urartians to the Persians, Byzantines, Seljuks, Mongols, Ottomans, and more, each leaving their mark on the architectural landscape. From the ancient temples of Garni and Geghard to the medieval monasteries of Haghpat and Sanahin, every structure tells a tale of resilience, innovation, and adaptation.
However, in recent times, several challenges have threatened Armenia’s architectural heritage. The devastating earthquake of 1988, the collapse of the Soviet Union, economic hardships, and urbanization have all posed significant obstacles to the preservation and revitalization of these architectural treasures.
Recognizing the importance of safeguarding this rich heritage, the LFA and the National Library of Armenia have joined forces to create the Armenian Pavilion at the 24th Triennale Milano International Exhibition. The pavilion aims to shed light on the architectural legacy of Armenia and to inspire a dialogue about the importance of historical preservation and adaptive reuse.
The exhibition will feature a curated selection of architectural projects that showcase the integration of traditional craftsmanship with contemporary design principles. Through this fusion of the old and the new, the Armenian Pavilion seeks to demonstrate how architectural heritage can be reimagined and revitalized in the modern world.
By referencing historical styles, materials, and techniques, Armenian architects and designers are achieving a delicate balance between preserving the past and embracing the future. The Armenian Pavilion will exemplify this approach, highlighting the innovative restoration projects, sustainable interventions, and adaptive reuse initiatives that are breathing life back into forgotten structures.
As we embark on this exciting journey, the Armenian Pavilion at the 24th Triennale Milano International Exhibition invites visitors to explore the intricate tapestry of Armenia’s architectural heritage and to reflect on the lessons it offers in the context of our rapidly evolving world. Through the lenses of the past, present, and future, we hope to inspire a renewed appreciation for the importance of architectural preservation, cultural identity, and sustainable design.
“Architecture should speak of its time and place, but yearn for timelessness.” – Frank Gehry
On May 13, 2025, the Library for Architecture (LFA), in collaboration with the National Library of Armenia, will inaugurate the Armenian Pavilion at the 24th Triennale Milano International Exhibition.
Read the original article
by jsendak | Apr 17, 2025 | DS Articles
[This article was first published on
R | Dr Tom Palmer, and kindly contributed to
R-bloggers]. (You can report issue about the content on this page
here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t.
Introduction
GitHub has recently announced that Windows ARM64 runners are
now available under the windows-11-arm label.
I help maintain an R package,
TwoSampleMR, which has quite alot of users. The package is not on CRAN because several of its dependencies are only on GitHub, and for a package to be on CRAN essentially all of its dependencies must also be on CRAN. As a result I am always interested to try installing the package on new operating systems and architectures.
(In this post I will use ARM and AARCH64 interchangeably.)
Setting up R AARCH64 on Windows on ARM
Avoiding confusion with the default runner software
It is important to mention that the x86_64 version of R 4.4.2 and RTools44 are included in the
default software set for the windows-latest GitHub Actions runner. And the directory including its binaries are on the PATH environment variable (specifically C:Program Files (x86)RR-4.4.2binx64). As a result if you run R, Rscript, or R CMD batch etc. in a shell in the runner you will obtain the x86_64 version of R (which runs under emulation on the ARM runner). Let’s say this is not what we want, so to setup the ARM version of R we need to install it ourselves.
Tomas Kalibera from the R Core Team has provided several excellent posts (
here and
here) about R for Windows on ARM, and installers for it have been available for some time.
The r-hub API does not yet provide the installer information for the AARCH64 version of R, so I came up with the following workflow file – amended from r-lib/actions to install R 4.5.0 and RTools45. Place such a (GitHub Actions workflow) file in a public GitHub repo in a .github/workflows directory, and enable GitHub Actions in the repo settings.
on:
push:
branches: [main, master]
pull_request:
branches: [main, master]
workflow_dispatch:
name: Check-install-win-11-arm
permissions: read-all
jobs:
windows-11-on-arm:
runs-on: windows-11-arm
name: windows-11-arm
strategy:
fail-fast: false
env:
GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
R_KEEP_PKG_SOURCE: yes
steps:
- name: Install R and RTools for Windows on ARM and install TwoSampleMR
run: |
$url = "https://www.r-project.org/nosvn/winutf8/aarch64/R-4-signed/R-4.5.0-aarch64.exe"
Invoke-WebRequest -Uri "$url" -OutFile R-4.5.0-aarch64.exe -UseBasicParsing -UserAgent "NativeHost"
Start-Process -FilePath R-4.5.0-aarch64.exe -ArgumentList "/install /norestart /verysilent /SUPPRESSMSGBOXES" -NoNewWindow -Wait
$url = "https://cran.r-project.org/bin/windows/Rtools/rtools45/files/rtools45-aarch64-6536-6492.exe"
Invoke-WebRequest -Uri "$url" -OutFile rtools45-aarch64-6536-6492.exe -UseBasicParsing -UserAgent "NativeHost"
Start-Process -FilePath rtools45-aarch64-6536-6492.exe -ArgumentList "/install /norestart /verysilent /SUPPRESSMSGBOXES" -NoNewWindow -Wait
$rscript = "C:Program FilesR-aarch64R-4.5.0binRscript.exe"
$arguments = "-e", "print(R.version); # the rest of your R code goes here ..."
& $rscript $arguments
Breaking down the final steps section of this;
- we define the url of the R 4.5.0 aarch64 installer;
- we then download the installer using
Invoke-WebRequest (note that the default shell in Windows is Powershell);
- we then run the installer using
Start-Process. I am not sure if I need all of the arguments I have specified here but it seems to work.
- We then do the same for RTools45.
- We then define a variable for the path to the Rscript.exe binary;
- we define a variable containing the arguments we want to pass to Rscript;
- we then invoke Rscript using our two variables and the
& call operator.
Then we navigate to our GitHub repo and view the output in the Actions tab under the relevant run.
Of course if you want to run your own R script you’ll need an initial step to checkout your repo.
To confirm that we really have launched the AARCH64 version of R we see the output of print(R.version) is as follows.
print(R.version)
#> _
#> platform aarch64-w64-mingw32
#> arch aarch64
#> os mingw32
#> crt ucrt
#> system aarch64, mingw32
#> status
#> major 4
#> minor 5.0
#> year 2025
#> month 04
#> day 11
#> svn rev 88135
#> language R
#> version.string R version 4.5.0 (2025-04-11 ucrt)
#> nickname How About a Twenty-Six
Summary
I have shown how to install the AARCH64 version of R and RTools45 on the recently released Windows on ARM runner in GitHub Actions.
As an aside, I note that we are now in the interesting position in that GitHub Actions now has Windows, macOS, and Ubuntu Linux all available on both x86_64 and ARM architectures.
Continue reading: Running R on Windows on ARM on GitHub Actions
Analysis of Running R on Windows on ARM on GitHub Actions
The original text discusses the new announcement by GitHub about the availability of Windows ARM64 runners under the label ‘windows-11-arm’. The author also provides a detailed guide about setting up R AARCH64 on Windows on ARM, with specific reference to maintaining the R package ‘TwoSampleMR’.
Long-term Implications and Future Developments
The development of R package ‘TwoSampleMR’ and its availability on new operating systems will greatly enhance its accessibility and usage amongst its extensive user base. The recent announcement by GitHub could signify that various platforms are beginning to accept and implement more powerful ARM64-based architectures for various software, including R. As ARM-based systems usually offer better power efficiency, this move can lead to energy savings and more efficient computing.
The implication is not just significant for the ‘TwoSampleMR’ package, but also for other software dependencies that rely on GitHub. It represents a step towards ensuring that these tools are compatible with newer systems and architectures.
In the long term, with ARM-based systems gaining more popularity and usage, it’s predictable that software and tools will increasingly become optimized for these systems.
Actionable Advice
For developers and IT professionals:
- Learning about ARM and AARCH64 architectures, and familiarising with installing software on these systems could become an essential skill. Training and development in this area would be beneficial.
- Developers maintaining R packages or similar tools should test their software on these new architectures to ensure compatibility and function.
- Developers should also consider updating their package dependencies on repositories like GitHub and CRAN, as this can help in broadening the user base of their packages.
For users and organizations:
- The migration to ARM64 architecture could bring about significant improvements in efficiency and power consumption. Organizations should therefore consider investing in this technology as a part of their long-term strategy.
- Scheduled check-ins for updates related to ARM64 support for various software tools will help avoid any potential compatibility issues.
Conclusion
This new development in ARM64 support for GitHub signifies a huge step forward in embracing more power-efficient architectures. This could potentially spark numerous advancements in software development and optimization for ARM-based systems. Both developers and users would benefit greatly by keeping abreast of such advancements and making necessary adjustments to adapt to these forward-looking changes.
Read the original article
