» Special Sessions

Fifty percent of the current activities in the US are technically automatable by adapting existing technologies. For example, each industrial robot in manufacturing replaces six workers. As a result, most individuals favor limits on replacing jobs through automation. While demand for physical and manual labor is declining, there is a huge demand for a workforce with social-emotional and technological skills.

The workforce of the future will have to be creative and innovative, rather than merely good at performing specific tasks. But the problem is that many individuals lack these skills, particularly if they suffer from cognitive disabilities or difficulties. Many argue that those skills are either innate or require extensive practice with human experts. Recent technical advancements have shown promise for technology to be able to understand the subtle nuances of soft behavioral skills and provide meaningful feedback to participants whenever they want, wherever they want. AI technologies, when designed appropriately, can successfully aid individuals in tasks such as job interviews, public speaking, negotiations, working as part of a team and even routine social interactions for people with autism. Can advances in affective computing potentially transform the future of workforce?

The goal of this special session is to encourage brave new ideas on using AI to improve human ability. In particular, we want to promote the following values as we innovate in this particular domain: 1) augment human beings rather than replace them; 2) be mindful of the potential misuse by recommending policy and social implications.

Topics of interest include but are not limited to:

Recognition

New techniques to recognize and quantify the subtle nuances of human soft skills.

Intervention

Evaluation of the new algorithms in the context of upskilling the workforce.

Applications

Use cases of designing new applications of affective computing in the context of future of work.

Feedback strategies

Designing new feedback strategies for improving human ability.

Policy and ethical framework

Social implications of advancing affective computing that understands human soft skills.

Behavioral health technologies are gaining increasing prominence in the delivery of mental health interventions. However, the affective elements underpinning mental health are not sufficiently addressed, despite the fact that the emotional component of mental health conditions is considerable, e.g.:

Affective problems are the main symptoms in some disorders.
Deficits in emotion regulation have been identified as significant factors in mental disorders.
The emotional impact of experience can be a key factor in triggering mental health problems.
The severe emotional burden takes its toll on both the patient and those around them.

Mental health affects the very core of who we are, and everyone’s experience of mental health disorders is unique. Approaches to mental health therefore need to take into account the diversity of lived experiences of mental illness. In addition, mental health issues do not impact uniformly across the population. It is therefore imperative that mental health services, through diagnosis, therapies and support, are strongly inclusive.

To address these issues, the aim of this special session is twofold:

To highlight the importance of emotions in many aspects of mental health, often neglected in models of mental disorders, and
To bring together different disciplines involved in the understanding and treatment of affect-related mental health, as well as different approaches and technologies within affective computing, to discuss how these different partners and aspects must work together to address mental health in an inclusive way.

With this double aim in mind, we solicit papers addressing the potential contributions of affective technologies and multi-disciplinary research to promote inclusive mental health, including topics such as (but not excluding other relevant topics):

In diagnosis and treatment, to assess patients and customize interventions.
Teaching or supporting emotion management and emotion regulation skills.
Affective companions and tools to reduce the emotional burden on patients and their carers.
Computational and embodied modeling of affect its involvement in cognition as a tool in psychopathology research.
Strengthening the inclusion of affective aspects of mental disorders in computational psychiatry.
Adaptive affective and interactive technologies: adapting to people with atypical affect (e.g., people with mental health disorders and non-neurotypical development.)

For further information, visit: http://www.emotion-modeling.info/MentalHealthACII2019.

Interactive robots are being deployed into homes, schools, shopping malls, hospitals, and retirement homes worldwide. Many of these robots have both the capacity to express emotions, as well as attempt to adapt to the emotions of the users around them, toward empathetic, compassionate interactions. It is clear, however, that affect is expressed in a plethora of ways, differing by culture, age, ability, social context, and more. Furthermore, social psychologists have emphasized the limitations of technology that uses basic emotions (joy, sadness, anger, fear, disgust, surprise) to analyze real-life situations. Roboticists in particular, burdened with real-time constraints and integration of many different skills onto one robot, often still use these older affective psychology paradigms, including off-the-shelf basic emotion detectors which work with limited populations and situations.

This special session on Affective Robotics for All highlights the opportunity for roboticists to challenge these older approaches and build new affective computing techniques, towards robots capable of affective interaction in real-life, diverse situations.

We will invite submissions on topics including, but not limited to:

Affective robotics for older adults
Affective robotics for youth and children
Affective robotics for people with disabilities
Affect models for cross-cultural or culture-specific interaction
Personalization techniques for affective robotics
Improved accuracy for real-time emotion perception on robots
Cognitive interactive robotics for affect learning and development
Techniques for real-time affect modeling
New models for affect recognition, expression and interaction
Affective robotics in the wild
Affective robots for education
Robot empathy
Robots for well-being

One aim of this special session is to understand the commonalities and dissimilarities between Affective Computing (AC) and other related areas, Kansei Engineering (KE), particularly in terms of culture and gender, for mutual development of both communities.

AC and KE are strongly overlapped. For example, both try to investigate their target constructs, which are not directly observable, by exploiting statistics/machine learning techniques with various types of measurements, e.g. behavior, physiological signals and subjective reports. However, there are also some differences. The central focus of AC is human emotions per se, while KE is oriented to improve product design by measuring human reactions induced by objects. Moreover, AC and KE originated in different cultures. This may be another cause of their differences since emotion is context-dependent. For example, the same arousal may be experienced as joy or anger depending on situational cues. Furthermore, context is usually shaped by culture (Munezero et al. 2014). Kansei is a Japanese term similar to the Western term of sentiment. Both imply an underlying attitude from people towards an entity. However, Western culture tends to separate logic reasoning (logos) and feelings (pathos), putting more importance on logos than pathos. On the other hand, Asians think about them as a unit, and even consider pathos more important. Therefore, it is important to understand whether cultural differences influence affective computing studies and how strongly.

Gender differences in emotion experiences and the interaction between gender and culture (Fischer & Manstead 2004) are also well known. However, in AC community compared to KE area, gender is often treated as a between-subject factor in many affective computing studies and detailed analysis has been avoided. For example, some KE researchers have been targeted a Japanese adjective kawaii which has a positive feminine meaning related to cute, lovable, and charming. However, kawaii objects are preferred not only by females but also by males, especially in young generations in Japan and some Asian countries. Kawaii baby and puppy are immature and incomplete, which causes empathy with them.

Clarifying such affective aspects caused by culture and gender in a special session can make significant contribution in the AC community. For example, it could help to realize universal user interfaces or computers that can understand and have culture- and gender-specific emotions.

Topics (Not limited to):

Affective computing from Asian perspective
Affective computing from Western perspective
Affective computing from Oceanian perspective
Affective computing from African perspective
Affective computing from Latin American perspective
Cross-cultural studies on affective computing
Gender studies on affective computing
Models of affective attributes of industrial products

Automatic detection of affect and emotion recognition in animals is a novel field that has gained popularity in recent years. The scientific study of animal expression of emotions started with the work of Charles Darwin (Darwin 1872). Analysing facial expressions of animals, gait analysis and modelling, and animal-human interaction are growing in significance recently as technology becomes an effective element in animal welfare. Studies discussing emotions in different animals and Interaction between human and animals have been sparsely presented in different conferences. There is increasing interest in this field for domesticated pets (such as cats and dogs), for livestock (such as cattle and sheep) and for wild animals. Our aim in this special session is to combine publications in the area of animal affect and interaction under one theme.

We invite submissions on topics including, but not limited to:

Computer vision for analysis of animals
Databases for animal behaviour analysis
Analysing animal behaviour
Animal gait analysis
Animal facial expressions analysis
Affect recognition in animals
Cross-species models of emotion
Modelling human-animal interaction
Modelling animal-machine interaction
Interfaces and tools for animal welfare

Psychological theories and evidence from multiple disciplines including neuroscience, economics and artificial intelligence suggest that the task of assigning reference-based values to subjective notions is better aligned with the underlying representations. An increasing number of studies have reported the advantages of ordinal annotation over alternative methods (e.g., nominal and interval descriptors) with respect to both reliability and validity. The impact of such evidence on affective computing is extremely important in the ways we annotate, analyze and process emotions. The emerging approach of using ordinal representations has led to improved performance across several tasks, including face analysis, speech recognition, body-based affective interaction, game applications, and retrieval of music and sounds.

Studies relying on the ordinal nature of emotions are scattered across different venues, including several conferences related to affective computing. This first ACII special session on the topic brings together researchers working on ordinal representation in affective computing.

We will invite submissions on topics including, but not limited to:

Psychological methods and tools for the ordinal representation of emotions
Statistical methods for ordinal label analysis
Preference learning and ranking-based methods for emotion recognition
Ordinal methods for the annotation of emotional behaviors
Ordinal multimodal corpora
Software for ordinal label processing

Special Sessions

Key Dates

197 Days

21 Hours

until Sep 3rd, 2019.

Newly added: