Pub Date : 2023-10-30DOI: 10.3389/fcomp.2023.1233905
Susan Loh, Yasu Santo, Marcus Foth
The current outsourcing of maintenance and the use of technological devices to automatically care for plants in buildings change the spatial experience between human office occupants and plants. This caretaker system distances people from plants, inclining us to regard them more as decorative objects. The relationship between humans and plants in a building is often unidirectional, with plants providing humans multiple benefits such as improved health and well-being, and increased worker productivity. In our human-plant interaction study, we developed a layer of care infrastructure within an office building that gives agency to people as a collective to interact with and take care of other non-human beings; that is, plants. In re-imagining mediated human-building interaction, we employed technology as an ambient mediator where people, plants and technology comprised the plant care system in a typical office building. A year-long design intervention was introduced within a typical office floor using artifacts (pots, shelves, and digital system) which we fabricated for the plants. From the results of an 8 week participation experiment together with data from qualitative interviews of 6 study participants, we identified five themes: Technology, Object/Thing, Infrastructuring, Commoning, and Care. Our analysis of these themes informs a care infrastructuring approach where both humans and plants become interdependent office co-inhabitants. By entangling with technology, care, and others, we present an infrastructuring layer to mediate human-building interactions with plants.
{"title":"Plant-human entanglements in buildings: designing for care infrastructuring with office occupants and pot plants","authors":"Susan Loh, Yasu Santo, Marcus Foth","doi":"10.3389/fcomp.2023.1233905","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1233905","url":null,"abstract":"The current outsourcing of maintenance and the use of technological devices to automatically care for plants in buildings change the spatial experience between human office occupants and plants. This caretaker system distances people from plants, inclining us to regard them more as decorative objects. The relationship between humans and plants in a building is often unidirectional, with plants providing humans multiple benefits such as improved health and well-being, and increased worker productivity. In our human-plant interaction study, we developed a layer of care infrastructure within an office building that gives agency to people as a collective to interact with and take care of other non-human beings; that is, plants. In re-imagining mediated human-building interaction, we employed technology as an ambient mediator where people, plants and technology comprised the plant care system in a typical office building. A year-long design intervention was introduced within a typical office floor using artifacts (pots, shelves, and digital system) which we fabricated for the plants. From the results of an 8 week participation experiment together with data from qualitative interviews of 6 study participants, we identified five themes: Technology, Object/Thing, Infrastructuring, Commoning, and Care. Our analysis of these themes informs a care infrastructuring approach where both humans and plants become interdependent office co-inhabitants. By entangling with technology, care, and others, we present an infrastructuring layer to mediate human-building interactions with plants.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136067720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-30DOI: 10.3389/fcomp.2023.1274832
Huma Jamil, Yajing Liu, Nathaniel Blanchard, Michael Kirby, Chris Peterson
In the ever-evolving landscape of deep learning, novel designs of neural network architectures have been thought to drive progress by enhancing embedded representations. However, recent findings reveal that the embedded representations of various state-of-the-art models are mappable to one another via a simple linear map, thus challenging the notion that architectural variations are meaningfully distinctive. While these linear maps have been established for traditional non-adversarial datasets, e.g., ImageNet, to our knowledge no work has explored the linear relation between adversarial image representations of these datasets generated by different CNNs. Accurately mapping adversarial images signals the feasibility of generalizing an adversarial defense optimized for a specific network. In this work, we demonstrate the existence of a linear mapping of adversarial inputs between different models that can be exploited to develop such model-agnostic, generalized adversarial defense. We further propose an experimental setup designed to underscore the concept of this model-agnostic defense. We train a linear classifier using both adversarial and non-adversarial embeddings within the defended space. Subsequently, we assess its performance using adversarial embeddings from other models that are mapped to this space. Our approach achieves an AUROC of up to 0.99 for both CIFAR-10 and ImageNet datasets.
{"title":"Leveraging linear mapping for model-agnostic adversarial defense","authors":"Huma Jamil, Yajing Liu, Nathaniel Blanchard, Michael Kirby, Chris Peterson","doi":"10.3389/fcomp.2023.1274832","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1274832","url":null,"abstract":"In the ever-evolving landscape of deep learning, novel designs of neural network architectures have been thought to drive progress by enhancing embedded representations. However, recent findings reveal that the embedded representations of various state-of-the-art models are mappable to one another via a simple linear map, thus challenging the notion that architectural variations are meaningfully distinctive. While these linear maps have been established for traditional non-adversarial datasets, e.g., ImageNet, to our knowledge no work has explored the linear relation between adversarial image representations of these datasets generated by different CNNs. Accurately mapping adversarial images signals the feasibility of generalizing an adversarial defense optimized for a specific network. In this work, we demonstrate the existence of a linear mapping of adversarial inputs between different models that can be exploited to develop such model-agnostic, generalized adversarial defense. We further propose an experimental setup designed to underscore the concept of this model-agnostic defense. We train a linear classifier using both adversarial and non-adversarial embeddings within the defended space. Subsequently, we assess its performance using adversarial embeddings from other models that are mapped to this space. Our approach achieves an AUROC of up to 0.99 for both CIFAR-10 and ImageNet datasets.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136105712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-25DOI: 10.3389/fcomp.2023.1154069
Jouni Pöyhönen, Martti Lehto
Global maritime transportation and logistics systems are essential parts of critical infrastructures in every society, and a crucial part of maritime logistics processes are seaports. In the coming years, digitalization and increased levels of autonomy in logistic transport chains are expected to take leaps forward. This development can help create safer, more efficient, more sustainable, and more reliable service chains to meet the requirements for a better quality of life and global prosperity. Port and harbor operations connect the maritime transport to other modes of transportation and enable multimodal transportation. Smart ports play a central role in future transport logistics and supply chains. Digitalization helps improve the efficiency of terminal systems in the processes of these ports. In the best cases, digitalization can also promote the reduction of emissions by optimizing port operations and enhancing cargo and people flows while improving the experience for all stakeholders. The improvement of port processes relies on the development of information and communication technology (ICT) as well as on industrial control systems (ICS) and operation technologies (OT). At the same time, the cyber security of maritime logistics also needs to be addressed. This article presents our findings related to the Sea4Value research goal on cyber security, which is a comprehensive cyber security architecture for port services at the system level. The article emphasizes the importance of a system of systems approach in terms of a comprehensive cyber security management process for port ecosystems. The description and recognition of management steps of every stakeholder are the key elements in this kind of process.
{"title":"Comprehensive cyber security for port and harbor ecosystems","authors":"Jouni Pöyhönen, Martti Lehto","doi":"10.3389/fcomp.2023.1154069","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1154069","url":null,"abstract":"Global maritime transportation and logistics systems are essential parts of critical infrastructures in every society, and a crucial part of maritime logistics processes are seaports. In the coming years, digitalization and increased levels of autonomy in logistic transport chains are expected to take leaps forward. This development can help create safer, more efficient, more sustainable, and more reliable service chains to meet the requirements for a better quality of life and global prosperity. Port and harbor operations connect the maritime transport to other modes of transportation and enable multimodal transportation. Smart ports play a central role in future transport logistics and supply chains. Digitalization helps improve the efficiency of terminal systems in the processes of these ports. In the best cases, digitalization can also promote the reduction of emissions by optimizing port operations and enhancing cargo and people flows while improving the experience for all stakeholders. The improvement of port processes relies on the development of information and communication technology (ICT) as well as on industrial control systems (ICS) and operation technologies (OT). At the same time, the cyber security of maritime logistics also needs to be addressed. This article presents our findings related to the Sea4Value research goal on cyber security, which is a comprehensive cyber security architecture for port services at the system level. The article emphasizes the importance of a system of systems approach in terms of a comprehensive cyber security management process for port ecosystems. The description and recognition of management steps of every stakeholder are the key elements in this kind of process.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135168231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-24DOI: 10.3389/fcomp.2023.1158476
Erivan Gonçalves Duarte, Isabelle Cossette, Marcelo M. Wanderley
Music educators and researchers have grown increasingly aware of the need for traditional musical practices to promote inclusive music for disabled people. Inclusive music participation has been addressed by Accessible Digital Musical Instruments (ADMIs), which welcome different ways of playing and perceiving music, with considerable impact on music-making for disabled people. ADMIs offer exciting possibilities for instrument design to consider and incorporate individual constraints (e.g., missing arm, low vision, hearing loss, etc.) more than traditional acoustic instruments, whose generally fixed design allows little room for disabled musicians inclusivity. Relatively few works discuss ADMIs in the context of disability studies, and no work has investigated the impact of different disability models in the process of designing inclusive music technology. This paper proposes criteria to classify ADMIs according to the medical, social, and cultural models of disability, then applies these criteria to evaluate eleven ADMIs targeting d/Deaf people. This analysis allows us to reflect on the design of ADMIs from different perspectives of disability, giving insights for future projects and deepening our understanding of medical, social, and cultural aspects of accessible music technology.
{"title":"Analysis of Accessible Digital Musical Instruments through the lens of disability models: a case study with instruments targeting d/Deaf people","authors":"Erivan Gonçalves Duarte, Isabelle Cossette, Marcelo M. Wanderley","doi":"10.3389/fcomp.2023.1158476","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1158476","url":null,"abstract":"Music educators and researchers have grown increasingly aware of the need for traditional musical practices to promote inclusive music for disabled people. Inclusive music participation has been addressed by Accessible Digital Musical Instruments (ADMIs), which welcome different ways of playing and perceiving music, with considerable impact on music-making for disabled people. ADMIs offer exciting possibilities for instrument design to consider and incorporate individual constraints (e.g., missing arm, low vision, hearing loss, etc.) more than traditional acoustic instruments, whose generally fixed design allows little room for disabled musicians inclusivity. Relatively few works discuss ADMIs in the context of disability studies, and no work has investigated the impact of different disability models in the process of designing inclusive music technology. This paper proposes criteria to classify ADMIs according to the medical, social, and cultural models of disability, then applies these criteria to evaluate eleven ADMIs targeting d/Deaf people. This analysis allows us to reflect on the design of ADMIs from different perspectives of disability, giving insights for future projects and deepening our understanding of medical, social, and cultural aspects of accessible music technology.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135265965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-23DOI: 10.3389/fcomp.2023.1209515
Sally Hamouda, Sahith Kancharla, Gurkirat Singh, Lin Yang, Zhuoqun Wang, Siliang Zhang, Raseen Nirjhar, John Golden
Data and visualizations are powerful tools that provide insights, analysis, and conclusions in a logical and easy-to-understand manner. However, the current school curriculum lacks adequate preparation for students to understand, analyze, interpret, or create complex data visualizations, which can hinder their potential careers in data science. To address this gap, our project aimed to develop a user-friendly web-based tool that provides interactive lessons on data and visualizations for elementary school children. The website consists of 12 lessons, categorized by grade levels (1st–2nd grade, 3rd–4th grade, and 5th–6th grade), and includes an interactive question-answer section. Users can scroll down after reading the lessons and practice questions based on the visualizations. The website also has the potential to incorporate games related to data and visualization. The lessons are implemented using React.js and Java with the Spring framework, and new lessons can easily be added by storing them in a markdown folder. The website features a navigation bar with tabs for Home, Lessons, Games, About, and Contact. Additionally, a feedback form is included to gather user feedback for further improvements. The website is currently in the testing stage, and future surveys for teachers and elementary school students will be added to enhance the features provided. Our study presents preliminary findings and serves as a foundational exploration. We acknowledge that further research and experimentation are required to validate and expand upon the results discussed herein.
{"title":"KiData: simple data visualization tool for future data scientists","authors":"Sally Hamouda, Sahith Kancharla, Gurkirat Singh, Lin Yang, Zhuoqun Wang, Siliang Zhang, Raseen Nirjhar, John Golden","doi":"10.3389/fcomp.2023.1209515","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1209515","url":null,"abstract":"Data and visualizations are powerful tools that provide insights, analysis, and conclusions in a logical and easy-to-understand manner. However, the current school curriculum lacks adequate preparation for students to understand, analyze, interpret, or create complex data visualizations, which can hinder their potential careers in data science. To address this gap, our project aimed to develop a user-friendly web-based tool that provides interactive lessons on data and visualizations for elementary school children. The website consists of 12 lessons, categorized by grade levels (1st–2nd grade, 3rd–4th grade, and 5th–6th grade), and includes an interactive question-answer section. Users can scroll down after reading the lessons and practice questions based on the visualizations. The website also has the potential to incorporate games related to data and visualization. The lessons are implemented using React.js and Java with the Spring framework, and new lessons can easily be added by storing them in a markdown folder. The website features a navigation bar with tabs for Home, Lessons, Games, About, and Contact. Additionally, a feedback form is included to gather user feedback for further improvements. The website is currently in the testing stage, and future surveys for teachers and elementary school students will be added to enhance the features provided. Our study presents preliminary findings and serves as a foundational exploration. We acknowledge that further research and experimentation are required to validate and expand upon the results discussed herein.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135367735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-20DOI: 10.3389/fcomp.2023.1191853
Shruthi Koratagere Anantha Kumar, Edward J. Oughton
How cost-efficient are potential infrastructure sharing business models for the 5G era (and beyond)? This significant question needs to be addressed if we are to deliver universal affordable broadband in line with Target 9.1 of the UN Sustainable Development Goals. Although almost two-thirds of the global population is now connected, many users still lack access to high-speed and reliable broadband connectivity. Indeed, some of the largest connectivity issues are associated with those living in areas of low economic viability. Consequently, this assessment evaluates the cost implications of different infrastructure sharing business models using a techno-economic assessment framework. The results indicate that a rural 5G neutral host network (NHN) strategy helps to reduce total cost between 10 and 50% compared with other sharing strategies. We also find that, compared to a baseline strategy with No Sharing , the net present value of rural 5G sharing strategies can earn between 30 and 90% more profit. The network upgrades to 5G using various sharing strategies are most sensitive to changes in the average revenue per user, the adoption rate, and the amount of existing site infrastructure. For example, the results from this study show that a 20% variation in demand revenue is estimated to increase the net present value of the sharing strategies by 2–5 times compared to the No Sharing strategy. Similarly, a 10% increase in existing infrastructure lowers the net present value by 8–30%. The infrastructure sharing strategies outlined in this study have the potential to enhance network viability while bridging the digital divide in remote and rural locations.
{"title":"Techno-economic assessment of 5G infrastructure sharing business models in rural areas","authors":"Shruthi Koratagere Anantha Kumar, Edward J. Oughton","doi":"10.3389/fcomp.2023.1191853","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1191853","url":null,"abstract":"How cost-efficient are potential infrastructure sharing business models for the 5G era (and beyond)? This significant question needs to be addressed if we are to deliver universal affordable broadband in line with Target 9.1 of the UN Sustainable Development Goals. Although almost two-thirds of the global population is now connected, many users still lack access to high-speed and reliable broadband connectivity. Indeed, some of the largest connectivity issues are associated with those living in areas of low economic viability. Consequently, this assessment evaluates the cost implications of different infrastructure sharing business models using a techno-economic assessment framework. The results indicate that a rural 5G neutral host network (NHN) strategy helps to reduce total cost between 10 and 50% compared with other sharing strategies. We also find that, compared to a baseline strategy with No Sharing , the net present value of rural 5G sharing strategies can earn between 30 and 90% more profit. The network upgrades to 5G using various sharing strategies are most sensitive to changes in the average revenue per user, the adoption rate, and the amount of existing site infrastructure. For example, the results from this study show that a 20% variation in demand revenue is estimated to increase the net present value of the sharing strategies by 2–5 times compared to the No Sharing strategy. Similarly, a 10% increase in existing infrastructure lowers the net present value by 8–30%. The infrastructure sharing strategies outlined in this study have the potential to enhance network viability while bridging the digital divide in remote and rural locations.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135617015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-19DOI: 10.3389/fcomp.2023.1281100
Phillip Kerger, Ryoji Miyazaki
We investigate a framework for binary image denoising via restricted Boltzmann machines (RBMs) that introduces a denoising objective in quadratic unconstrained binary optimization (QUBO) form well-suited for quantum annealing. The denoising objective is attained by balancing the distribution learned by a trained RBM with a penalty term for derivations from the noisy image. We derive the statistically optimal choice of the penalty parameter assuming the target distribution has been well-approximated, and further suggest an empirically supported modification to make the method robust to that idealistic assumption. We also show under additional assumptions that the denoised images attained by our method are, in expectation, strictly closer to the noise-free images than the noisy images are. While we frame the model as an image denoising model, it can be applied to any binary data. As the QUBO formulation is well-suited for implementation on quantum annealers, we test the model on a D-Wave Advantage machine, and also test on data too large for current quantum annealers by approximating QUBO solutions through classical heuristics.
{"title":"Quantum image denoising: a framework via Boltzmann machines, QUBO, and quantum annealing","authors":"Phillip Kerger, Ryoji Miyazaki","doi":"10.3389/fcomp.2023.1281100","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1281100","url":null,"abstract":"We investigate a framework for binary image denoising via restricted Boltzmann machines (RBMs) that introduces a denoising objective in quadratic unconstrained binary optimization (QUBO) form well-suited for quantum annealing. The denoising objective is attained by balancing the distribution learned by a trained RBM with a penalty term for derivations from the noisy image. We derive the statistically optimal choice of the penalty parameter assuming the target distribution has been well-approximated, and further suggest an empirically supported modification to make the method robust to that idealistic assumption. We also show under additional assumptions that the denoised images attained by our method are, in expectation, strictly closer to the noise-free images than the noisy images are. While we frame the model as an image denoising model, it can be applied to any binary data. As the QUBO formulation is well-suited for implementation on quantum annealers, we test the model on a D-Wave Advantage machine, and also test on data too large for current quantum annealers by approximating QUBO solutions through classical heuristics.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135730414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-16DOI: 10.3389/fcomp.2023.1154737
David Kirsh
In this article, I consider how scientific theories may explain architectural atmosphere. Architects use atmosphere to refer to a holistic, emergent property of a space that partly determines the mood of inhabitants. It is said to be a “subtle, intangible, ambient quality of a place” that also significantly shapes the way we interact with a space. It is caused by the way light, texture, materials, layout, geometry, acoustics, smell, and other perceptual properties influence affect. But it goes beyond these individually because of non-linear interactions between them. In sections one and two, I explain what an externalist account of the atmosphere would look like. This is an interpretation that objectifies the atmosphere, treating it as a complex causal property of buildings and spaces, accessible to scientific study through ethnographic research, through quantifying and minutely observing and recording humans and the buildings they are in, and then using machine learning and statistical analyses to identify correlations. The goal is to push the identification of the underlying external attributes as far as possible, ultimately to where a machine might enter a room, move around, and then label its atmosphere. In section three, I explore an internalist or subjectivist account of the atmosphere. This is the position that pushes back on machine identification of atmospheres. A subjectivist interpretation is harder to study scientifically because it involves knowing so much about the inner state and the history of a person. Culture, incoming mood, prior experience and associations, interests, tasks, social interaction, and more may all affect mood. Section four explores the frequently underestimated role—on emotion and space comprehension—played by the tasks that occupants perform while in a space, and the way their surrounding social and technological context intrudes on their encounter. I introduce and defend the view that tasks, social context, and nearby technology situate a person in a different environment than when they are inactive. This complicates the search for atmosphere. Nonetheless, I end on an optimistic note that there may yet be a place for atmosphere in the neuroscience of architecture, but it will be much different than our current thinking.
{"title":"Atmosphere, mood, and scientific explanation","authors":"David Kirsh","doi":"10.3389/fcomp.2023.1154737","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1154737","url":null,"abstract":"In this article, I consider how scientific theories may explain architectural atmosphere. Architects use atmosphere to refer to a holistic, emergent property of a space that partly determines the mood of inhabitants. It is said to be a “subtle, intangible, ambient quality of a place” that also significantly shapes the way we interact with a space. It is caused by the way light, texture, materials, layout, geometry, acoustics, smell, and other perceptual properties influence affect. But it goes beyond these individually because of non-linear interactions between them. In sections one and two, I explain what an externalist account of the atmosphere would look like. This is an interpretation that objectifies the atmosphere, treating it as a complex causal property of buildings and spaces, accessible to scientific study through ethnographic research, through quantifying and minutely observing and recording humans and the buildings they are in, and then using machine learning and statistical analyses to identify correlations. The goal is to push the identification of the underlying external attributes as far as possible, ultimately to where a machine might enter a room, move around, and then label its atmosphere. In section three, I explore an internalist or subjectivist account of the atmosphere. This is the position that pushes back on machine identification of atmospheres. A subjectivist interpretation is harder to study scientifically because it involves knowing so much about the inner state and the history of a person. Culture, incoming mood, prior experience and associations, interests, tasks, social interaction, and more may all affect mood. Section four explores the frequently underestimated role—on emotion and space comprehension—played by the tasks that occupants perform while in a space, and the way their surrounding social and technological context intrudes on their encounter. I introduce and defend the view that tasks, social context, and nearby technology situate a person in a different environment than when they are inactive. This complicates the search for atmosphere. Nonetheless, I end on an optimistic note that there may yet be a place for atmosphere in the neuroscience of architecture, but it will be much different than our current thinking.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136116738","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-16DOI: 10.3389/fcomp.2023.1286591
Ilmo Salmenperä, Jukka K. Nurminen
Restricted Boltzmann machines are common machine learning models that can utilize quantum annealing devices in their training processes as quantum samplers. While this approach has shown promise as an alternative to classical sampling methods, the limitations of quantum annealing hardware, such as the number of qubits and the lack of connectivity between the qubits, still pose a barrier to wide-scale adoption. We propose the use of multiple software techniques such as dropout method, passive labeling, and parallelization techniques for addressing these hardware limitations. The study found that using these techniques along with quantum sampling showed comparable results to its classical counterparts in certain contexts, while in others the increased complexity of the sampling process hindered the performance of the trained models. This means that further research into the behavior of quantum sampling needs to be done to apply quantum annealing to training tasks of more complicated RBM models.
{"title":"Software techniques for training restricted Boltzmann machines on size-constrained quantum annealing hardware","authors":"Ilmo Salmenperä, Jukka K. Nurminen","doi":"10.3389/fcomp.2023.1286591","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1286591","url":null,"abstract":"Restricted Boltzmann machines are common machine learning models that can utilize quantum annealing devices in their training processes as quantum samplers. While this approach has shown promise as an alternative to classical sampling methods, the limitations of quantum annealing hardware, such as the number of qubits and the lack of connectivity between the qubits, still pose a barrier to wide-scale adoption. We propose the use of multiple software techniques such as dropout method, passive labeling, and parallelization techniques for addressing these hardware limitations. The study found that using these techniques along with quantum sampling showed comparable results to its classical counterparts in certain contexts, while in others the increased complexity of the sampling process hindered the performance of the trained models. This means that further research into the behavior of quantum sampling needs to be done to apply quantum annealing to training tasks of more complicated RBM models.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136113788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-16DOI: 10.3389/fcomp.2023.1274928
Ngo Duong Ha, Nhu Y. Tran, Le Nhi Lam Thuy, Ikuko Shimizu, Pham The Bao
Classification of school violence has been proven to be an effective solution for preventing violence within educational institutions. As a result, technical proposals aimed at enhancing the efficacy of violence classification are of considerable interest to researchers. This study explores the utilization of the SORT tracking method for localizing and tracking objects in videos related to school violence, coupled with the application of LSTM and GRU methods to enhance the accuracy of the violence classification model. Furthermore, we introduce the concept of a padding box to localize, identify actions, and recover tracked objects lost during video playback. The integration of these techniques offers a robust and efficient system for analyzing and preventing violence in educational environments. The results demonstrate that object localization and recovery algorithms yield improved violent classification outcomes compared to both the SORT tracking and violence classification algorithms alone, achieving an impressive accuracy rate of 72.13%. These experimental findings hold promise, especially in educational settings, where the assumption of camera stability is justifiable. This distinction is crucial due to the unique characteristics of violence in educational environments, setting it apart from other forms of violence.
{"title":"Violence region localization in video and the school violent actions classification","authors":"Ngo Duong Ha, Nhu Y. Tran, Le Nhi Lam Thuy, Ikuko Shimizu, Pham The Bao","doi":"10.3389/fcomp.2023.1274928","DOIUrl":"https://doi.org/10.3389/fcomp.2023.1274928","url":null,"abstract":"Classification of school violence has been proven to be an effective solution for preventing violence within educational institutions. As a result, technical proposals aimed at enhancing the efficacy of violence classification are of considerable interest to researchers. This study explores the utilization of the SORT tracking method for localizing and tracking objects in videos related to school violence, coupled with the application of LSTM and GRU methods to enhance the accuracy of the violence classification model. Furthermore, we introduce the concept of a padding box to localize, identify actions, and recover tracked objects lost during video playback. The integration of these techniques offers a robust and efficient system for analyzing and preventing violence in educational environments. The results demonstrate that object localization and recovery algorithms yield improved violent classification outcomes compared to both the SORT tracking and violence classification algorithms alone, achieving an impressive accuracy rate of 72.13%. These experimental findings hold promise, especially in educational settings, where the assumption of camera stability is justifiable. This distinction is crucial due to the unique characteristics of violence in educational environments, setting it apart from other forms of violence.","PeriodicalId":52823,"journal":{"name":"Frontiers in Computer Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136113796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：