by jsendak | Feb 26, 2024 | Computer Science
arXiv:2402.14947v1 Announce Type: cross
Abstract: Governments use propaganda, including through visual content — or Politically Salient Image Patterns (PSIP) — on social media, to influence and manipulate public opinion. In the present work, we collected Telegram post-history of from 989 Russian milbloggers to better understand the social and political narratives that circulated online in the months surrounding Russia’s 2022 full-scale invasion of Ukraine. Overall, we found an 8,925% increase (p<0.001) in the number of posts and a 5,352% increase (p<0.001) in the number of images posted by these accounts in the two weeks prior to the invasion. We also observed a similar increase in the number and intensity of politically salient manipulated images that circulated on Telegram. Although this paper does not evaluate malice or coordination in these activities, we do conclude with a call for further research into the role that manipulated visual media has in the lead-up to instability events and armed conflict.
Expert Commentary: Analyzing Politically Salient Image Patterns (PSIP) on Social Media
Introduction
In today’s digital age, social media has become a powerful platform for governments and political entities to disseminate their messages and shape public opinion. A recent study titled “Politically Salient Image Patterns (PSIP) on Social Media: A Case Study of the Russian Invasion of Ukraine” focuses on the use of visual content or PSIP on Telegram, a popular messaging app. Through the analysis of data from 989 Russian milbloggers, the study aims to shed light on the social and political narratives that emerged surrounding Russia’s invasion of Ukraine in 2022.
Understanding the Impact of Visual Content
The study highlights the significant role played by visual content in influencing public opinion. Images possess the power to evoke emotions, convey messages, and shape narratives more effectively than text alone. As such, understanding the patterns and messages within PSIP is crucial in comprehending the impact of propaganda on social media platforms.
Multi-Disciplinary Nature of PSIP Analysis
An analysis of PSIP involves a multidisciplinary approach, encompassing various fields such as multimedia information systems, animations, artificial reality, augmented reality, and virtual realities.
- Multimedia Information Systems: The study relies on data collection and analysis methodologies used in the field of multimedia information systems. By examining post-history on Telegram, researchers gain insights into the types of visual content employed by milbloggers.
- Animations: The use of animated images, GIFs, or short videos can be a powerful method to captivate and influence viewers. Analyzing PSIP can reveal if such techniques were utilized during Russia’s invasion of Ukraine.
- Artificial Reality and Virtual Realities: The manipulation of visual content can extend beyond the physical realm through the use of augmented reality or virtual reality technologies. By examining PSIP, researchers can assess whether such technologies were harnessed to amplify propaganda efforts.
Implications and Future Directions
The findings of this study provide valuable insights into the role of PSIP in shaping social and political narratives during Russia’s invasion of Ukraine. Understanding the techniques employed by governments and political entities in manipulating public opinion is crucial for safeguarding the integrity of digital platforms and democracy at large.
Furthermore, this research opens up avenues for future studies. By expanding the dataset to include different regions or conflicts, researchers can compare the strategies employed and identify common patterns or approaches utilized in different geopolitical contexts.
In conclusion, the analysis of Politically Salient Image Patterns (PSIP) on social media offers a deeper understanding of the influence of visual content in shaping public opinion. The interdisciplinary nature of this analysis connects various fields and provides a comprehensive perspective on the dynamics at play in this digital age.
Read the original article
by jsendak | Feb 26, 2024 | Computer Science
The article provides a comprehensive overview of the Bin Packing Problem, highlighting its significance in discrete optimization and its relevance to real-world problems. It acknowledges that various theoretical and practical tools have been used to address this problem, with the most effective approaches being based on Linear Programming. Furthermore, it mentions how Constraint Programming can be valuable when the Bin Packing Problem is part of a larger problem.
One interesting aspect addressed in this work is the exploration of how GPUs (Graphics Processing Units) can enhance the propagation algorithm of the Bin Packing constraint. The article presents two approaches motivated by knapsack reasoning and alternative lower bounds, respectively. It is crucial to mention that GPUs are known for their high parallel processing power, which makes them potentially suitable for improving the performance of certain algorithms.
By evaluating the implementations of these GPU-accelerated approaches, the research team compares them to state-of-the-art techniques on different benchmarks from the literature. The results obtained suggest that the GPU-accelerated lower bounds offer a promising alternative for tackling large instances of the Bin Packing Problem.
This study contributes to the field of discrete optimization by introducing GPU-accelerated techniques for enhancing the Bin Packing constraint’s propagation algorithm. By leveraging the parallel processing capabilities of GPUs, these approaches show potential for significantly improving the efficiency and scalability of solving large instances of the problem.
In terms of future developments, it would be interesting to see how these GPU-accelerated techniques could be further optimized and extended. Additionally, it would be valuable to explore their applicability to other optimization problems and investigate how different problem characteristics may influence their effectiveness.
Read the original article
by jsendak | Feb 24, 2024 | Computer Science
Analyzing Prediction Markets and Their Limitations
Prediction markets have proven to be valuable tools for estimating probabilities of claims that can be resolved at a specific point in time. These markets excel in predicting uncertainties related to real-world events and even values of primitive recursive functions. However, their direct application to questions without a fixed resolution criterion is challenging, leading to predictions about whether a sentence will be proven rather than its truth.
When it comes to questions that lack a fixed resolution criterion, a different approach is necessary. Such questions often involve countable unions or intersections of more basic events or are represented as First-Order-Logic sentences on the Arithmetical Hierarchy. In more complex cases, they may even transcend First-Order Logic and fall into the realm of hyperarithmetical sentences.
In this paper, the authors propose an alternative approach to betting on events without a fixed resolution criterion using options. These options can be viewed as bets on the outcome of a “verification-falsification game,” offering a new framework for addressing logical uncertainty. This work stands in contrast to the existing framework of Garrabrant induction and aligns with the constructivist stance in the philosophy of mathematics.
By introducing the concept of options in prediction markets, this research has far-reaching implications for both philosophy and mathematical logic. It provides a fresh perspective on addressing uncertainties in a broader range of questions and challenges the traditional methods by offering an alternative framework that accommodates events lacking fixed resolution criteria. These findings encourage further exploration and could lead to significant advancements in our understanding and utilization of prediction markets.
Read the original article
by jsendak | Feb 23, 2024 | Computer Science
arXiv:2402.14326v1 Announce Type: new
Abstract: Offloading computing to edge servers is a promising solution to support growing video understanding applications at resource-constrained IoT devices. Recent efforts have been made to enhance the scalability of such systems by reducing inference costs on edge servers. However, existing research is not directly applicable to pixel-level vision tasks such as video semantic segmentation (VSS), partly due to the fluctuating VSS accuracy and segment bitrate caused by the dynamic video content. In response, we present Penance, a new edge inference cost reduction framework. By exploiting softmax outputs of VSS models and the prediction mechanism of H.264/AVC codecs, Penance optimizes model selection and compression settings to minimize the inference cost while meeting the required accuracy within the available bandwidth constraints. We implement Penance in a commercial IoT device with only CPUs. Experimental results show that Penance consumes a negligible 6.8% more computation resources than the optimal strategy while satisfying accuracy and bandwidth constraints with a low failure rate.
Analysis of Penance: Edge Inference Cost Reduction Framework
In this article, the authors introduce Penance, a new framework for reducing edge inference costs in video semantic segmentation (VSS) tasks. With the growing demand for video understanding applications on resource-constrained IoT devices, offloading computing to edge servers has become a promising solution. However, existing research is not directly applicable to pixel-level vision tasks like VSS, mainly due to the dynamic nature of video content, which leads to fluctuating accuracy and segment bitrate.
Penance addresses this challenge by leveraging the softmax outputs of VSS models and the prediction mechanism of H.264/AVC codecs. By optimizing model selection and compression settings, Penance aims to minimize the inference cost while meeting the required accuracy within the available bandwidth constraints. It is worth noting that Penance is implemented on a commercial IoT device with only CPUs, making it accessible to a wide range of devices.
The multi-disciplinary nature of this work is evident in its integration of computer vision (specifically VSS), video codecs (H.264/AVC), and edge computing. It combines knowledge from these diverse domains to develop a novel solution that addresses the specific challenges faced in edge inference for VSS.
When considering the wider field of multimedia information systems, Penance contributes to the efficiency and scalability of video understanding applications on IoT devices. By reducing inference costs at the edge, it enables resource-constrained devices to perform complex vision tasks like semantic segmentation without relying heavily on cloud resources. This can lead to improved response times, reduced latency, and increased privacy.
Furthermore, Penance has relevance to various aspects of multimedia technologies such as animations, artificial reality, augmented reality, and virtual realities. These technologies often involve real-time video processing and analysis, where efficient edge inference is crucial for a seamless and immersive user experience. By optimizing inference costs, Penance can support the delivery of rich multimedia content in these applications without compromising on performance.
In conclusion, Penance is an innovative framework that addresses the challenges of edge inference for video semantic segmentation tasks. Its integration of various technologies and its impact on the wider field of multimedia information systems, animations, artificial reality, augmented reality, and virtual realities make it a significant contribution to the advancement of edge computing in the context of video understanding applications.
Read the original article
by jsendak | Feb 23, 2024 | Computer Science
Existing approaches to Theory of Mind (ToM) in Artificial Intelligence (AI) have predominantly focused on prompted or cue-based ToM. While this has yielded useful insights into how AI systems can infer and reason about the mental states of others, there is a growing consensus that this approach may limit the development of Artificial Social Intelligence (ASI).
In this article, we propose an alternative perspective by introducing the concept of spontaneous ToM. Spontaneous ToM refers to the ability of AI systems to reason about others’ mental states using unintentional and possibly uncontrollable cognitive functions. By grounding social reasoning in these natural, automatic processes, AI systems can achieve a more human-like understanding of the minds of others.
The Limitations of Prompted ToM
Prompted ToM, as commonly explored in AI research, involves explicitly providing cues or prompts to an AI system to elicit inferences about the mental states of others. While this approach has proven valuable, it presents certain limitations.
Firstly, prompted ToM relies heavily on external cues, which may not always be available in real-world social interactions. This limits the generalizability and applicability of prompted ToM models. Spontaneous ToM, on the other hand, leverages cognitive functions that operate automatically and can be applied in a broader range of situations.
Secondly, prompted ToM may neglect the important role of unconscious mental processes in social reasoning. By exclusively focusing on explicit prompts, AI systems miss out on the rich and nuanced information that can be gleaned from spontaneous cognitive processes. Incorporating spontaneous ToM allows for a more comprehensive understanding of others’ mental states.
A Principled Approach to AI ToM
We advocate for a principled approach to studying and developing AI ToM, which involves considering both prompted and spontaneous ToM. By combining these two forms of social reasoning, AI systems can exhibit a robust and generalized ASI.
Principled AI ToM would require research efforts to explore the cognitive mechanisms underlying spontaneous ToM. Understanding these innate processes can help AI systems mimic human-like social reasoning and enhance their ability to predict and respond to the mental states of others.
Furthermore, developing AI systems with spontaneous ToM could have significant implications for various applications. For instance, in human-robot interaction, AI systems with spontaneous ToM could anticipate the intentions and needs of users more effectively, leading to more seamless and intuitive interactions.
The Future of ASI
The integration of spontaneous ToM into AI systems marks an exciting direction for the field of ASI. As researchers delve further into understanding the cognitive processes involved in spontaneous social reasoning, we can expect significant advancements in the development of AI systems that are genuinely capable of understanding and interacting with humans in a social context.
However, challenges lie ahead. Unintentional cognitive functions can be challenging to model, as they are often implicit and difficult to define explicitly. Overcoming these challenges will require interdisciplinary collaborations between computer science, cognitive science, neuroscience, and related disciplines.
In summary, by moving beyond prompted ToM and embracing spontaneous social reasoning, AI researchers can unlock the full potential of ASI. As we continue to investigate the intricacies of human cognition, we can lay the groundwork for AI systems that possess a genuine understanding of others’ mental states.
Read the original article
by jsendak | Feb 22, 2024 | Computer Science
arXiv:2304.11783v2 Announce Type: replace-cross
Abstract: Rip currents pose a significant danger to those who visit beaches, as they can swiftly pull swimmers away from shore. Detecting these currents currently relies on costly equipment and is challenging to implement on a larger scale. The advent of unmanned aerial vehicles (UAVs) and camera technology, however, has made monitoring near-shore regions more accessible and scalable. This paper proposes a new framework for detecting rip currents using video-based methods that leverage optical flow estimation, offshore direction calculation, earth camera projection with almost local-isometric embedding on the sphere, and temporal data fusion techniques. Through the analysis of videos from multiple beaches, including Palm Beach, Haulover, Ocean Reef Park, and South Beach, as well as YouTube footage, we demonstrate the efficacy of our approach, which aligns with human experts’ annotations.
The Multi-Disciplinary Nature of Rip Current Detection
Rip current detection is a complex problem that requires a multi-disciplinary approach to tackle effectively. In this research paper, the authors propose a new framework that combines concepts from computer vision, signal processing, and geographical mapping to detect rip currents using video-based methods.
The use of unmanned aerial vehicles (UAVs) and camera technology enables the monitoring of near-shore regions in a more accessible and scalable manner. By analyzing videos from multiple beaches and leveraging techniques such as optical flow estimation, offshore direction calculation, earth camera projection, and temporal data fusion, the proposed framework aims to improve rip current detection accuracy.
One of the key components of this framework is optical flow estimation, which involves tracking the motion of objects in a video sequence. By analyzing the flow patterns in the video, it becomes possible to identify regions where rip currents are likely to occur. This technique has been widely used in computer vision applications, but its adaptation for rip current detection is novel and promising.
In addition to optical flow estimation, the framework also incorporates offshore direction calculation. This involves determining the direction in which rip currents are flowing, which is crucial for accurately predicting their behavior. By combining information from multiple cameras positioned at different angles, the framework can estimate the offshore direction with higher precision.
To further enhance the accuracy of rip current detection, the proposed framework leverages earth camera projection with almost local-isometric embedding on the sphere. This technique allows for better representation of the spatial relationships between different regions of interest in the video, enabling more accurate detection and tracking of rip currents.
Integration with Multimedia Information Systems
The research presented in this paper highlights the integration of multimedia information systems with rip current detection. By leveraging video-based methods and analyzing footage from multiple sources, including YouTube, the framework expands the scope of available data for analysis. This integration with multimedia information systems enables a broader understanding of rip current patterns and behaviors, leading to more accurate detection and prediction.
Applications in Artificial Reality, Augmented Reality, and Virtual Realities
The proposed framework for rip current detection using video-based methods has significant implications for artificial reality, augmented reality, and virtual realities. By accurately detecting and predicting rip currents, this technology can be utilized to create immersive virtual environments that simulate real-world beach conditions.
For example, virtual reality simulations could provide training scenarios for lifeguards, allowing them to practice rescue operations in a safe and controlled environment. Augmented reality applications could also enhance beach safety by overlaying real-time rip current information on smartphone screens or heads-up displays, providing beachgoers with crucial alerts and guidance.
Furthermore, the integration of rip current detection technology with artificial reality, augmented reality, and virtual realities could enable novel experiences for users. Imagine a virtual beach experience where users can witness the power and danger of rip currents firsthand, providing valuable educational opportunities and promoting beach safety awareness.
Conclusion
The proposed framework for rip current detection using video-based methods demonstrates the power of a multi-disciplinary approach. By combining concepts from computer vision, signal processing, and geographical mapping, the framework aims to improve the accuracy and scalability of rip current monitoring.
The integration of multimedia information systems, animations, artificial reality, augmented reality, and virtual realities opens up new possibilities for enhancing beach safety, training lifeguards, and creating immersive experiences. The utilization of unmanned aerial vehicles (UAVs) and camera technology will continue to play a vital role in advancing the field of rip current detection and enhancing our understanding of coastal dynamics.
Read the original article
