Search

Robots deployed in public settings enter spaces that humans live and work in. Studies of HRI in public tend to prioritise direct and deliberate interactions. Yet this misses the most common form of response to robots, which ranges from subtle fleeting interactions to virtually ignoring them. Taking an ethnomethodological approach building on video recordings, we show how robots become embedded in urban spaces both from a perspective of the social assembly of the physical environment (the streetscape) and the socially organised nature of everyday street life. We show how such robots are effectively ‘granted passage’ through these spaces as a result of the practical work of the streets’ human inhabitants. We detail the contingent nature of the streetscape, drawing attention to its various members and the accommodation work they are doing. We demonstrate the importance of studying robots during their whole deployment, and approaches that focus on members’ interactional work., Funding Agencies|Engineering and Physical Sciences Research Council [EP/T022493/1, MMW 2020.0086], WASP-HS AI in Motion

Robots deployed in public settings enter spaces that humans live and work in. Studies of HRI in public tend to prioritise direct and deliberate interactions. Yet this misses the most common form of response to robots, which ranges from subtle fleeting interactions to virtually ignoring them. Taking an ethnomethodological approach building on video recordings, we show how robots become embedded in urban spaces both from a perspective of the social assembly of the physical environment (the streetscape) and the socially organised nature of everyday street life. We show how such robots are effectively ‘granted passage’ through these spaces as a result of the practical work of the streets’ human inhabitants. We detail the contingent nature of the streetscape, drawing attention to its various members and the accommodation work they are doing. We demonstrate the importance of studying robots during their whole deployment, and approaches that focus on members’ interactional work., Funding Agencies|Engineering and Physical Sciences Research Council [EP/T022493/1, MMW 2020.0086], WASP-HS AI in Motion

Robots deployed in public settings enter spaces that humans live and work in. Studies of HRI in public tend to prioritise direct and deliberate interactions. Yet this misses the most common form of response to robots, which ranges from subtle fleeting interactions to virtually ignoring them. Taking an ethnomethodological approach building on video recordings, we show how robots become embedded in urban spaces both from a perspective of the social assembly of the physical environment (the streetscape) and the socially organised nature of everyday street life. We show how such robots are effectively ‘granted passage’ through these spaces as a result of the practical work of the streets’ human inhabitants. We detail the contingent nature of the streetscape, drawing attention to its various members and the accommodation work they are doing. We demonstrate the importance of studying robots during their whole deployment, and approaches that focus on members’ interactional work., Funding Agencies|Engineering and Physical Sciences Research Council [EP/T022493/1, MMW 2020.0086], WASP-HS AI in Motion

Robots that can talk and move may turn from tools to potential participants, which poses new methodological challenges, particularly for transcription. This chapter first presents best practices for transcribing multimodal robot actions, focusing on sound. Robots animate the action repertoires that they are given by their designers and can do so again and again, producing virtually identical sequences. This work discusses how to transcribe such repeated action, balancing between the general script and situated moves. Moving from transcription to analysis, the chapter pays special attention to differences in how humans and robots demonstrate understanding of sequential actions. The chapter closes by demonstrating how transcription can reveal the dynamic character of robot participation, which is often assisted and scaffolded by humans who frame the robot's actions as relevant and accountable.

Robots deployed in public settings enter spaces that humans live and work in. Studies of HRI in public tend to prioritise direct and deliberate interactions. Yet this misses the most common form of response to robots, which ranges from subtle fleeting interactions to virtually ignoring them. Taking an ethnomethodological approach building on video recordings, we show how robots become embedded in urban spaces both from a perspective of the social assembly of the physical environment (the streetscape) and the socially organised nature of everyday street life. We show how such robots are effectively ‘granted passage’ through these spaces as a result of the practical work of the streets’ human inhabitants. We detail the contingent nature of the streetscape, drawing attention to its various members and the accommodation work they are doing. We demonstrate the importance of studying robots during their whole deployment, and approaches that focus on members’ interactional work., Funding Agencies|Engineering and Physical Sciences Research Council [EP/T022493/1, MMW 2020.0086], WASP-HS AI in Motion

Robots deployed in public settings enter spaces that humans live and work in. Studies of HRI in public tend to prioritise direct and deliberate interactions. Yet this misses the most common form of response to robots, which ranges from subtle fleeting interactions to virtually ignoring them. Taking an ethnomethodological approach building on video recordings, we show how robots become embedded in urban spaces both from a perspective of the social assembly of the physical environment (the streetscape) and the socially organised nature of everyday street life. We show how such robots are effectively ‘granted passage’ through these spaces as a result of the practical work of the streets’ human inhabitants. We detail the contingent nature of the streetscape, drawing attention to its various members and the accommodation work they are doing. We demonstrate the importance of studying robots during their whole deployment, and approaches that focus on members’ interactional work., WASP-HS AI in Motion

Federated learning (FL) is an emerging machine learning (ML) training paradigm where clients own their data and collaborate to train a global model, without revealing any data to the server and other participants. Researchers commonly perform experiments in a simulation environment to quickly iterate on ideas. However, existing open-source tools do not offer the efficiency required to simulate FL on larger and more realistic FL datasets. We introduce pfl-research, a fast, modular, and easy-to-use Python framework for simulating FL. It supports TensorFlow, PyTorch, and non-neural network models, and is tightly integrated with state-of-the-art privacy algorithms. We study the speed of open-source FL frameworks and show that pfl-research is 7-72$\times$ faster than alternative open-source frameworks on common cross-device setups. Such speedup will significantly boost the productivity of the FL research community and enable testing hypotheses on realistic FL datasets that were previously too resource intensive. We release a suite of benchmarks that evaluates an algorithm's overall performance on a diverse set of realistic scenarios. The code is available on GitHub at https://github.com/apple/pfl-research.

In this chapter, the authors convey some of the main ways the term ‘salutogenesis’ is used today. Antonovsky introduced the term salutogenesis in his 1979 and 1987 books on the Salutogenic Model of Health, but salutogenesis subsequently has come to refer to a core concept in the model in particular; the sense of coherence. This usage has advanced to the point that some writers have coined the term ‘sense of coherence theory’. The term salutogenesis is also frequently used to refer, more generally, to an approach to health theory, research and practice emphasising resources that people may call on to improve health.

Settings are defined by the World Health Organization (1998) as “the place or social context in which people engage in daily activities in which environmental, organizational, and personal factors interact to affect health and well-being.” Such settings range from small-scale home/family to (international) organizations and large cities and thus differ in size, in their degree of formalized organization and their relationships to society. The chapters in Part V review how salutogenesis has been applied to health promotion research and practice in a broad range of settings: organizations in general, schools, higher education, workplace, military settings, neighborhood/communities, cities, and restorative environments. The following synthesis demonstrates that applying salutogenesis to various settings and linking salutogenesis with other models established in these settings has the great potential to generate ideas on how to advance the general salutogenic model.

In this chapter, the author provides the reader with a useful description of Salutogenesis meeting places. These are the Society for Theory and Research on Salutogenesis (STARS) and the Global Working Group on Salutogenesis (GWG-Sal) of the International Union for Health Promotion and Education (IUHPE). The author also provides information about the Center on Salutogenesis at the University of Zürich in Switzerland. The Center is the host organization of both meeting places. The key message of this chapter is that one’s involvement in STARS will extend one’s professional network, open new avenues for research and publishing, and help achieve a vision of “salutogenesis for thriving societies.” The reader wanting to connect more directly to a global salutogenesis network will find this chapter to be of great practical value.

Salutogenesis Theory and Methods: Developments, Innovations and Next Steps: This chapter is this handbook’s ‘grand finale’, addressing the potential of salutogenesis to nurture thriving societies. The handbook’s editors reflect on the advancement of salutogenesis concerning theory development, applying the salutogenic model in community settings and helping society tackle crises such as COVID-19.

This chapter presents models, measures, and intervention approaches that relate to the double nature of work and its salutogenic quality. Hereby, the view of Aaron Antonovsky is enhanced insofar that health-promoting, salutogenic job characteristics are not solely understood as mitigating the pathogenic effects of stressors at work but have a distinct effect on positive health outcomes. In the chapter, Antonovsky’s original model is first specified and simplified for the context of work. Next, Antonovsky’s line of thinking is related to frameworks researching job resources and demands. After a review of the prevalence of salutogenic measures in worksite health promotion, the point of making salutogenesis more visible in work-related research and practice is elaborated. This is illustrated with a practical example of a survey-feedback process promoting salutogenic work.

Organisations influence the health of society through three major paths: the health of their employees through working conditions, the health of their customers through the quality of their products or services and the population’s health at large through their socio-ecological impact. This chapter focuses on the first path of organisations’ impact on employee health through working conditions. It complements the chapter on salutogenic work by expanding the level of analysis to organisational characteristics. The chapter aims to be particularly applicable to for-profit organisations, in which it is exceptionally challenging to introduce a health agenda.

In this chapter, the authors convey some of the main ways the term ‘salutogenesis’ is used today. Antonovsky introduced the term salutogenesis in his 1979 and 1987 books on the Salutogenic Model of Health, but salutogenesis subsequently has come to refer to a core concept in the model in particular; the sense of coherence. This usage has advanced to the point that some writers have coined the term ‘sense of coherence theory’. The term salutogenesis is also frequently used to refer, more generally, to an approach to health theory, research and practice emphasising resources that people may call on to improve health.

In this chapter, the author provides the reader with a useful description of Salutogenesis meeting places. These are the Society for Theory and Research on Salutogenesis (STARS) and the Global Working Group on Salutogenesis (GWG-Sal) of the International Union for Health Promotion and Education (IUHPE). The author also provides information about the Center on Salutogenesis at the University of Zürich in Switzerland. The Center is the host organization of both meeting places. The key message of this chapter is that one’s involvement in STARS will extend one’s professional network, open new avenues for research and publishing, and help achieve a vision of “salutogenesis for thriving societies.” The reader wanting to connect more directly to a global salutogenesis network will find this chapter to be of great practical value.

Organisations influence the health of society through three major paths: the health of their employees through working conditions, the health of their customers through the quality of their products or services and the population’s health at large through their socio-ecological impact. This chapter focuses on the first path of organisations’ impact on employee health through working conditions. It complements the chapter on salutogenic work by expanding the level of analysis to organisational characteristics. The chapter aims to be particularly applicable to for-profit organisations, in which it is exceptionally challenging to introduce a health agenda.

This chapter presents models, measures, and intervention approaches that relate to the double nature of work and its salutogenic quality. Hereby, the view of Aaron Antonovsky is enhanced insofar that health-promoting, salutogenic job characteristics are not solely understood as mitigating the pathogenic effects of stressors at work but have a distinct effect on positive health outcomes. In the chapter, Antonovsky’s original model is first specified and simplified for the context of work. Next, Antonovsky’s line of thinking is related to frameworks researching job resources and demands. After a review of the prevalence of salutogenic measures in worksite health promotion, the point of making salutogenesis more visible in work-related research and practice is elaborated. This is illustrated with a practical example of a survey-feedback process promoting salutogenic work.

Salutogenesis Theory and Methods: Developments, Innovations and Next Steps: This chapter is this handbook’s ‘grand finale’, addressing the potential of salutogenesis to nurture thriving societies. The handbook’s editors reflect on the advancement of salutogenesis concerning theory development, applying the salutogenic model in community settings and helping society tackle crises such as COVID-19.

Revised and updated from the first edition are background and historical chapters that trace the development of the salutogenic model of health and flesh out the central concepts, most notably generalized resistance resources and the sense of coherence that differentiate salutogenesis from pathogenesis. From there, experts describe a range of real-world applications within and outside health contexts. Many new chapters emphasize intervention research findings. Readers will find numerous practical examples of how to implement salutogenesis to enhance the health and well-being of families, infants and young children, adolescents, unemployed young people, pre-retirement adults, and older people. A dedicated section addresses how salutogenesis helps tackle vulnerability, with chapters on at-risk children, migrants, prisoners, emergency workers, and disaster-stricken communities. Wide-ranging coverage includes new topics beyond health, like intergroup conflict, politics and policy-making, and architecture. The book also focuses on applying salutogenesis in birth and neonatal care clinics, hospitals and primary care, schools and universities, workplaces, and towns and cities. A special section focuses on developments in salutogenesis methods and theory.

Settings are defined by the World Health Organization (1998) as “the place or social context in which people engage in daily activities in which environmental, organizational, and personal factors interact to affect health and well-being.” Such settings range from small-scale home/family to (international) organizations and large cities and thus differ in size, in their degree of formalized organization and their relationships to society. The chapters in Part V review how salutogenesis has been applied to health promotion research and practice in a broad range of settings: organizations in general, schools, higher education, workplace, military settings, neighborhood/communities, cities, and restorative environments. The following synthesis demonstrates that applying salutogenesis to various settings and linking salutogenesis with other models established in these settings has the great potential to generate ideas on how to advance the general salutogenic model.

Robots naturally emit sound, but we still know little about how sound can serve as an interface that makes a robot’s behavior explainable to humans. This dissertation draws on insights about human practices for coordinating bodily activities through sound, investigating how they could inform robot design. My work builds on three video corpora, involving i) a Cozmo robot in ten family homes, ii) autonomous public shuttle buses in an urban environment, and iii) a teamwork robot prototype controlled by a researcher and interacting with study participants in an experimental setting. I approached the data from two methodological angles, exploring how they can speak to each other: I first carried out an empirical analysis of the video data from an Ethnomethodology and Conversation Analysis (EMCA) perspective, focusing on how humans make sense of robot sound on a moment-by-moment basis in naturally occurring interaction. Subsequently, taking an Interaction Design perspective, I used my video recordings as a design material for exploring how robot sound could be designed in and for real-time interaction. My work contributes to Human-Robot Interaction through detailed studies of robots in the world (rather than in the lab), focusing on how participants make sense of robot sounds. I present a novel framework for designing sound in and for interaction and a prototyping practice that allows practitioners to embed an EMCA stance into their designs. The dissertation contributes to EMCA by describing how members embed autonomous machines into the social organization of activities and how humans treat robots as participants in the interaction. I make a contribution to the development of EMCA hybrid studies by seeking a synthesis between EMCA and robot interaction design., Trots att ljud är en naturlig del av en robots närvaro vet vi fortvarande väldigt lite om hur ljud kan användas i gränssnitt för att göra robotars beteende förståeligt för människor. Denna avhandling utgår från nya insikter om hur människor använder sina röster i kroppsliga aktiviteter, för att undersöka hur denna kunskap kan användas vid gestaltning av robotar. Avhandlingen bygger på tre videokorpusar som visar i) leksaksroboten Cozmo i tio olika barnfamiljers hemmiljö, ii) två autonoma bussar i stadsmiljö och iii) en forskarstyrd prototyp av en robot för grupparbete i en experimentell miljö. Korpusarna studerades utifrån ett etnometodologiskt och interaktionsanalytiskt perspektiv (eng. ethnomethodology and conversation analysis, EMCA). Analysen fokuserade på hur människor visar sin förståelse av robotljud i naturligt förekommande interaktioner. De två sistnämnda korpusarna användes dessutom som material för interaktionsdesign i syfte att utforska hur robotljud kan utformas för att stödja realtidsinteraktion. Arbetet bidrar till fältet människa-robotinteraktion genom att erbjuda detaljerade studier av robotar i världen (i motsats till laboratoriemiljö) med fokus på hur deltagare i sampel med en robot förstår dess ljud. Avhandlingen föreslår ett nytt ramverk för att utforma ljud för interaktionella syften och en metod för att implementera ett EMCA-förhållningssätt inom designpraktiker. Arbetet beskriver även hur autonoma maskiner kan ingå i socialt organiserade aktiviteter och hur robotar kan behandlas som deltagare i interaktion med människor. Slutligen bidrar avhandlingen även till utvecklingen av EMCA-hybridstudier genom att utforska möjligheten att utveckla en EMCA-informerad metod för design av robotinteraktion.

Sound is an important interaction modality in human interaction, which robot design is only starting to tap into. Drawing on insights about how human sounds support coordination of bodily activities, this work focuses on how robots can communicate through sound in concrete interactions in the wild. My work contributes a focus on how users make sense of sound in everyday interaction, promotes re-consideration of what HRI is designing for and stimulates the development of new HRI design methods., Funding Agencies|Swedish Research Council [2016-00827]

Wie wird man ‚Literaturpapst‘? Die Antwort lautet im frühen 18. Jahrhundert im Grunde genommen wie heute: Man musste Aufmerksamkeit auf sich ziehen. Dem Kampf um diese Aufmerksamkeit widmete Gottsched beträchtliche Energie, die in unzählige literaturkritische Zeitschriftenbeiträge floss. Rezensieren und kritisieren, Stellung beziehen zu literarischen und gelehrten Veröffentlichungen, über Literatur streiten, und das durchaus polemisch – darin fand Gottsched eines seiner auch für sein Selbstverständnis wesentlichsten Tätigkeitsfelder. Literaturtheoretiker und Literaturkritiker waren für Gottsched nur in Personalunion denkbar. Dem literaturtheoretischen Selbstbekenntnis zu Rationalität entspricht dabei der Anspruch des Kritikers auf letztinstanzliche Autorität über die von ihm beurteilten Werke. Kann dieses Zusammenspiel von Theorie und Kritik funktionieren? Indem der Artikel diese Frage verfolgt, wird das Verhältnis von Gottscheds theoretischem Programm der Kritik zu seiner kritischen Praxis untersucht.

Delays in the completion of joint actions are sometimes unavoidable. How should a robot communicate that it cannot immediately act or respond in a collaborative task? Drawing on video recordings of a face scanning activity in family homes, we investigate how humans make sense of a Cozmo robot’s delays on a moment-by-moment basis. Cozmo’s sounds and embodied actions are recognized as indicators of delay but encourage human participants to act in ways that undermine the scanning process. In comparing the robot’s delay management strategies with human-human vocal and embodied practices, we demonstrate key differences in the sequences that impact how the robot is understood. The study demonstrates how delay events are accomplished as embodied displays that are distributed across co-participants. We present a framework for making delay transparent through situated explanations, particularly in the form of non-lexical sounds and bodily actions., Funding: Swedish Research Council [2016-00827]

Horns and sirens are important tools for communicating on the road, which are still understudied in autonomous vehicles. While HRI has explored different ways in which robots could sound, we focus on the range of actions that a single sound can accomplish in interaction. In a Research through Design study involving autonomous shuttle buses in public transport, we explored sound design with the help of voice-overs to video recordings of the buses on the road and Wizard-of-Oz tests in live traffic. The buses are slowed down by (unnecessary) braking in response to people getting close. We found that prolonged jingles draw attention to the bus and invite interaction, while repeated short beeps and bell sounds can instruct the movement of others away from the bus. We highlight the importance of designing sound in sequential interaction and describe a new method for embedding video interaction analysis in the design process.

Robots naturally emit sound, but we still know little about how sound can serve as an interface that makes a robot’s behavior explainable to humans. This dissertation draws on insights about human practices for coordinating bodily activities through sound, investigating how they could inform robot design. My work builds on three video corpora, involving i) a Cozmo robot in ten family homes, ii) autonomous public shuttle buses in an urban environment, and iii) a teamwork robot prototype controlled by a researcher and interacting with study participants in an experimental setting. I approached the data from two methodological angles, exploring how they can speak to each other: I first carried out an empirical analysis of the video data from an Ethnomethodology and Conversation Analysis (EMCA) perspective, focusing on how humans make sense of robot sound on a moment-by-moment basis in naturally occurring interaction. Subsequently, taking an Interaction Design perspective, I used my video recordings as a design material for exploring how robot sound could be designed in and for real-time interaction. My work contributes to Human-Robot Interaction through detailed studies of robots in the world (rather than in the lab), focusing on how participants make sense of robot sounds. I present a novel framework for designing sound in and for interaction and a prototyping practice that allows practitioners to embed an EMCA stance into their designs. The dissertation contributes to EMCA by describing how members embed autonomous machines into the social organization of activities and how humans treat robots as participants in the interaction. I make a contribution to the development of EMCA hybrid studies by seeking a synthesis between EMCA and robot interaction design., Trots att ljud är en naturlig del av en robots närvaro vet vi fortvarande väldigt lite om hur ljud kan användas i gränssnitt för att göra robotars beteende förståeligt för människor. Denna avhandling utgår från nya insikter om hur människor använder sina röster i kroppsliga aktiviteter, för att undersöka hur denna kunskap kan användas vid gestaltning av robotar. Avhandlingen bygger på tre videokorpusar som visar i) leksaksroboten Cozmo i tio olika barnfamiljers hemmiljö, ii) två autonoma bussar i stadsmiljö och iii) en forskarstyrd prototyp av en robot för grupparbete i en experimentell miljö. Korpusarna studerades utifrån ett etnometodologiskt och interaktionsanalytiskt perspektiv (eng. ethnomethodology and conversation analysis, EMCA). Analysen fokuserade på hur människor visar sin förståelse av robotljud i naturligt förekommande interaktioner. De två sistnämnda korpusarna användes dessutom som material för interaktionsdesign i syfte att utforska hur robotljud kan utformas för att stödja realtidsinteraktion. Arbetet bidrar till fältet människa-robotinteraktion genom att erbjuda detaljerade studier av robotar i världen (i motsats till laboratoriemiljö) med fokus på hur deltagare i sampel med en robot förstår dess ljud. Avhandlingen föreslår ett nytt ramverk för att utforma ljud för interaktionella syften och en metod för att implementera ett EMCA-förhållningssätt inom designpraktiker. Arbetet beskriver även hur autonoma maskiner kan ingå i socialt organiserade aktiviteter och hur robotar kan behandlas som deltagare i interaktion med människor. Slutligen bidrar avhandlingen även till utvecklingen av EMCA-hybridstudier genom att utforska möjligheten att utveckla en EMCA-informerad metod för design av robotinteraktion.

Robots naturally emit sound, but we still know little about how sound can serve as an interface that makes a robot’s behavior explainable to humans. This dissertation draws on insights about human practices for coordinating bodily activities through sound, investigating how they could inform robot design. My work builds on three video corpora, involving i) a Cozmo robot in ten family homes, ii) autonomous public shuttle buses in an urban environment, and iii) a teamwork robot prototype controlled by a researcher and interacting with study participants in an experimental setting. I approached the data from two methodological angles, exploring how they can speak to each other: I first carried out an empirical analysis of the video data from an Ethnomethodology and Conversation Analysis (EMCA) perspective, focusing on how humans make sense of robot sound on a moment-by-moment basis in naturally occurring interaction. Subsequently, taking an Interaction Design perspective, I used my video recordings as a design material for exploring how robot sound could be designed in and for real-time interaction. My work contributes to Human-Robot Interaction through detailed studies of robots in the world (rather than in the lab), focusing on how participants make sense of robot sounds. I present a novel framework for designing sound in and for interaction and a prototyping practice that allows practitioners to embed an EMCA stance into their designs. The dissertation contributes to EMCA by describing how members embed autonomous machines into the social organization of activities and how humans treat robots as participants in the interaction. I make a contribution to the development of EMCA hybrid studies by seeking a synthesis between EMCA and robot interaction design., Trots att ljud är en naturlig del av en robots närvaro vet vi fortvarande väldigt lite om hur ljud kan användas i gränssnitt för att göra robotars beteende förståeligt för människor. Denna avhandling utgår från nya insikter om hur människor använder sina röster i kroppsliga aktiviteter, för att undersöka hur denna kunskap kan användas vid gestaltning av robotar. Avhandlingen bygger på tre videokorpusar som visar i) leksaksroboten Cozmo i tio olika barnfamiljers hemmiljö, ii) två autonoma bussar i stadsmiljö och iii) en forskarstyrd prototyp av en robot för grupparbete i en experimentell miljö. Korpusarna studerades utifrån ett etnometodologiskt och interaktionsanalytiskt perspektiv (eng. ethnomethodology and conversation analysis, EMCA). Analysen fokuserade på hur människor visar sin förståelse av robotljud i naturligt förekommande interaktioner. De två sistnämnda korpusarna användes dessutom som material för interaktionsdesign i syfte att utforska hur robotljud kan utformas för att stödja realtidsinteraktion. Arbetet bidrar till fältet människa-robotinteraktion genom att erbjuda detaljerade studier av robotar i världen (i motsats till laboratoriemiljö) med fokus på hur deltagare i sampel med en robot förstår dess ljud. Avhandlingen föreslår ett nytt ramverk för att utforma ljud för interaktionella syften och en metod för att implementera ett EMCA-förhållningssätt inom designpraktiker. Arbetet beskriver även hur autonoma maskiner kan ingå i socialt organiserade aktiviteter och hur robotar kan behandlas som deltagare i interaktion med människor. Slutligen bidrar avhandlingen även till utvecklingen av EMCA-hybridstudier genom att utforska möjligheten att utveckla en EMCA-informerad metod för design av robotinteraktion.

Robots naturally emit sound, but we still know little about how sound can serve as an interface that makes a robot’s behavior explainable to humans. This dissertation draws on insights about human practices for coordinating bodily activities through sound, investigating how they could inform robot design. My work builds on three video corpora, involving i) a Cozmo robot in ten family homes, ii) autonomous public shuttle buses in an urban environment, and iii) a teamwork robot prototype controlled by a researcher and interacting with study participants in an experimental setting. I approached the data from two methodological angles, exploring how they can speak to each other: I first carried out an empirical analysis of the video data from an Ethnomethodology and Conversation Analysis (EMCA) perspective, focusing on how humans make sense of robot sound on a moment-by-moment basis in naturally occurring interaction. Subsequently, taking an Interaction Design perspective, I used my video recordings as a design material for exploring how robot sound could be designed in and for real-time interaction. My work contributes to Human-Robot Interaction through detailed studies of robots in the world (rather than in the lab), focusing on how participants make sense of robot sounds. I present a novel framework for designing sound in and for interaction and a prototyping practice that allows practitioners to embed an EMCA stance into their designs. The dissertation contributes to EMCA by describing how members embed autonomous machines into the social organization of activities and how humans treat robots as participants in the interaction. I make a contribution to the development of EMCA hybrid studies by seeking a synthesis between EMCA and robot interaction design., Trots att ljud är en naturlig del av en robots närvaro vet vi fortvarande väldigt lite om hur ljud kan användas i gränssnitt för att göra robotars beteende förståeligt för människor. Denna avhandling utgår från nya insikter om hur människor använder sina röster i kroppsliga aktiviteter, för att undersöka hur denna kunskap kan användas vid gestaltning av robotar. Avhandlingen bygger på tre videokorpusar som visar i) leksaksroboten Cozmo i tio olika barnfamiljers hemmiljö, ii) två autonoma bussar i stadsmiljö och iii) en forskarstyrd prototyp av en robot för grupparbete i en experimentell miljö. Korpusarna studerades utifrån ett etnometodologiskt och interaktionsanalytiskt perspektiv (eng. ethnomethodology and conversation analysis, EMCA). Analysen fokuserade på hur människor visar sin förståelse av robotljud i naturligt förekommande interaktioner. De två sistnämnda korpusarna användes dessutom som material för interaktionsdesign i syfte att utforska hur robotljud kan utformas för att stödja realtidsinteraktion. Arbetet bidrar till fältet människa-robotinteraktion genom att erbjuda detaljerade studier av robotar i världen (i motsats till laboratoriemiljö) med fokus på hur deltagare i sampel med en robot förstår dess ljud. Avhandlingen föreslår ett nytt ramverk för att utforma ljud för interaktionella syften och en metod för att implementera ett EMCA-förhållningssätt inom designpraktiker. Arbetet beskriver även hur autonoma maskiner kan ingå i socialt organiserade aktiviteter och hur robotar kan behandlas som deltagare i interaktion med människor. Slutligen bidrar avhandlingen även till utvecklingen av EMCA-hybridstudier genom att utforska möjligheten att utveckla en EMCA-informerad metod för design av robotinteraktion.

Horns and sirens are important tools for communicating on the road, which are still understudied in autonomous vehicles. While HRI has explored different ways in which robots could sound, we focus on the range of actions that a single sound can accomplish in interaction. In a Research through Design study involving autonomous shuttle buses in public transport, we explored sound design with the help of voice-overs to video recordings of the buses on the road and Wizard-of-Oz tests in live traffic. The buses are slowed down by (unnecessary) braking in response to people getting close. We found that prolonged jingles draw attention to the bus and invite interaction, while repeated short beeps and bell sounds can instruct the movement of others away from the bus. We highlight the importance of designing sound in sequential interaction and describe a new method for embedding video interaction analysis in the design process.

When aiming to design for intuitive interaction, a good understanding of hu- man behavior is essential. In this chapter we dive into studies on how humans use vocalizations and prosody in everyday interaction. Contrasting six ex- amples from human-human and human-robot interaction, we highlight how insights on human practices can inform the design of robot sound in interac- tion. We present three main lessons, demonstrating that a) both human vo- calizations and robot sound are semantically underspecified, b) human sound production is embodied, and robot sound should therefore be analyzed and designed multimodally, and c) sound can be easily adapted for complex par- ticipation.

n/a

While federated learning (FL) has recently emerged as a promising approach to train machine learning models, it is limited to only preliminary explorations in the domain of automatic speech recognition (ASR). Moreover, FL does not inherently guarantee user privacy and requires the use of differential privacy (DP) for robust privacy guarantees. However, we are not aware of prior work on applying DP to FL for ASR. In this paper, we aim to bridge this research gap by formulating an ASR benchmark for FL with DP and establishing the first baselines. First, we extend the existing research on FL for ASR by exploring different aspects of recent $\textit{large end-to-end transformer models}$: architecture design, seed models, data heterogeneity, domain shift, and impact of cohort size. With a $\textit{practical}$ number of central aggregations we are able to train $\textbf{FL models}$ that are \textbf{nearly optimal} even with heterogeneous data, a seed model from another domain, or no pre-trained seed model. Second, we apply DP to FL for ASR, which is non-trivial since DP noise severely affects model training, especially for large transformer models, due to highly imbalanced gradients in the attention block. We counteract the adverse effect of DP noise by reviving per-layer clipping and explaining why its effect is more apparent in our case than in the prior work. Remarkably, we achieve user-level ($7.2$, $10^{-9}$)-$\textbf{DP}$ (resp. ($4.5$, $10^{-9}$)-$\textbf{DP}$) with a 1.3% (resp. 4.6%) absolute drop in the word error rate for extrapolation to high (resp. low) population scale for $\textbf{FL with DP in ASR}$., Comment: Under review

In this paper, we start by training End-to-End Automatic Speech Recognition (ASR) models using Federated Learning (FL) and examining the fundamental considerations that can be pivotal in minimizing the performance gap in terms of word error rate between models trained using FL versus their centralized counterpart. Specifically, we study the effect of (i) adaptive optimizers, (ii) loss characteristics via altering Connectionist Temporal Classification (CTC) weight, (iii) model initialization through seed start, (iv) carrying over modeling setup from experiences in centralized training to FL, e.g., pre-layer or post-layer normalization, and (v) FL-specific hyperparameters, such as number of local epochs, client sampling size, and learning rate scheduler, specifically for ASR under heterogeneous data distribution. We shed light on how some optimizers work better than others via inducing smoothness. We also summarize the applicability of algorithms, trends, and propose best practices from prior works in FL (in general) toward End-to-End ASR models., Comment: In Proceedings of the IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) 2023

Federated learning (FL) combined with differential privacy (DP) offers machine learning (ML) training with distributed devices and with a formal privacy guarantee. With a large population of devices, FL with DP produces a performant model in a timely manner. However, for applications with a smaller population, not only does the model utility degrade as the DP noise is inversely proportional to population, but also the training latency increases since waiting for enough clients to become available from a smaller pool is slower. In this work, we thus propose expanding the population based on domain adaptation techniques to speed up the training and improves the final model quality when training with small populations. We empirically demonstrate that our techniques can improve the utility by 13% to 30% on real-world language modeling datasets.

This chapter is this handbook’s ‘grand finale’, addressing the potential of salutogenesis to nurture thriving societies. The handbook’s editors reflect on the advancement of salutogenesis concerning theory development, applying the salutogenic model in community settings and helping society tackle crises such as COVID-19.