Workshop on Social and Affective Intelligence


The inaugural Workshop on Social and Affective Intelligence (SAI) at ACII 2023 invites researchers from emotion science, social cognition, affective computing, HMI, and HRI to discuss cross-disciplinary perspectives on social and affective intelligence in humans and machines. Social Intelligence includes processes and competencies for perceiving, representing, reasoning about, and participating in social interactions. Affective Intelligence includes processes and competencies for communicating and managing emotions, feelings, moods, and other affective phenomena. This workshop will include keynote talks, paper presentations, and group brainstorming discussions focused on the following areas: (1) What are the relationships between social intelligence and affective intelligence in human cognition? How can insights from these relationships in humans influence modeling approaches in affective computing, HMI, and HRI? What are best practices for designing virtual and embodied agents that can jointly sense, perceive, interpret, and simulate social and affective phenomena? The SAI workshop at ACII 2023 will offer an interdisciplinary forum to strengthen connections between emotion science and research towards social and affective intelligence in machines.

Workshop Website:

Workshop Chair: Leena Mathur

Workshop on Addressing Social Context in Affective Computing (ASOCA)


Intelligent systems are expected to have empathic abilities for various tasks in social settings, such as tutoring and elderly care. Affective Computing (AC) has made progress in this area but has yet to address the highly context-sensitive nature of human cognitive-affective processing in social settings. The specifics of different social situations or the people involved in them (e.g., their norms and values) can strongly influence how emotions are expressed through behavior and what emotional expressions (or responses to them) are considered appropriate. As such, AC systems may easily fail to function correctly in diverse real-world situations due to their inability to consider variations in social context, resulting in inaccurate inferences and inappropriate behavior.

Our workshop seeks to establish an interdisciplinary platform for discussing research on Social Context in AC, promoting joint research projects, exchanging methods, and critically evaluating current and future efforts. We welcome technical, empirical (incl. case studies), and theoretical contributions (incl. position papers) that stimulate discussions on (1) what aspects of social context AC systems should consider, (2) how these systems might be enabled to do so, and (3) how research on context-sensitive AC systems might enable a better understanding of human cognitive-affective processing.

Workshop Website:

Workshop Chair: Bernd Dudzik

AHRI 2023: Second Workshop on Affective Human-Robot Interaction


In recent years, robotic applications have seen an increasing real-world deployment. It is common in these applications that a user interacts directly with a robot. In such Human-Robot Interaction (HRI), trust and mutual adaptation is established and maintained through a positive social relationship between the robot and the human interactor, and relies on the perceived competence of a robot on the social-emotional dimension. How a user perceives a robot’s social intelligence and their social relationship with the robot can have a direct influence on the outcomes of an HRI system. Moreover, in many HRI applications, social-emotional interaction with the intended users is the main goal of the system or a core strategy to achieve the desired outcomes. Such affective HRI applications require emotion-awareness and social-emotional competence in the robot’s functions to deliver acceptable services.

Following the success in 2022 (, the second AHRI workshop will continue to provide a communication and collaboration platform for researchers working on affective computing, HRI, social robotics, and AI and robotics application. In alignment with the ACII 2023’s theme in “Affective Computing: Context and Multimodality”, we especially welcome submissions on HRI in multimodal and naturalistic interaction contexts. This workshop will focus on discussing the following topics:

1. How to adaptively/accurately perceive unimodal or multimodal affective human behaviour in HRI under particular interaction context.

2. How to efficiently generate natural and affective robot behaviour in HRI that is appropriate for the interaction context. 

3. How to measure the benefits and outcomes of affective HRI applications with a user-centred and contextualised approach.

Workshop Website:

Workshop Chair: Leimin Tian

EPiC 2023: The Emotion Physiology and Experience Collaboration


The debate on physiological traces of emotions is still open. This competition-based workshop aims to further the debate by examining how well continuous ratings of valence and arousal can be modeled using physiological features. The competition and workshop are part of a larger proposed initiative – the Emotion Physiology and Experience Collaboration (EPiC) – which aims to accelerate both theoretical and applied research on the relationship between emotions and physiology. During the workshop, we will discuss the outcomes of the competition as well as proposals for next steps in EPiC roadmap, particularly the largest-ever multicultural study on emotion physiology and experience.

Workshop Website:

Workshop Chair: Stanisław Saganowski

LiLAC: Lifelong Learning in Affective Computing (Canceled)


Despite state-of-the-art performance on affect recognition benchmarks, deep learning solutions struggle in real-world applications where systems dynamically interact with different users. Assuming a clear separation of training and test settings requires models to have all the data available apriori, making such methods unsuitable or, at the least, inefficient when data distributions shift with each user or task. Lifelong or Continual Learning (CL) aims to address this challenge, enabling systems to adapt with a continuous and sequential stream of data, acquired from non-stationary or changing environments. This makes CL a ‘natural fit’ for real-world affective computing. This workshop aims to initiate a systematic and structured discussion towards the successful adoption of the CL paradigm for affective computing, bringing together a multidisciplinary group of researchers to:

i) start formalising key challenges towards dynamic adaptation in affect perception models;

ii) bring forth the merits and demerits of existing CL-based works in the context of affective computing;

iii) investigate novel methodologies adopting the CL paradigm for affective computing;

iv) formalise the benchmarks and evaluative metrics for CL-based affective computing.

Workshop Website:

Workshop Chair: Nikhil Churamani

 mWELL: Affective Computing for Mental Wellbeing: Challenges, Opportunities, and Promising Synergies

The prevalence of mental illness is globally on the rise, leading to high individual and societal burdens. Many people in need do not seek professional help, the ones that do are confronted with limited availability of practitioners, and despite considerable progress, the success rate of therapies remains limited. While more and more digital health solutions are emerging, their adoption and reported success are still low.

This workshop aims to bring together researchers in Affective Computing (AC), clinicians in the emerging area of digital mental health and digital psychiatry, developers from industry, and policymakers to discuss what aspects of digital mental health tools can most benefit from AC technologies and existing technologies already incorporating AC, such as embodied conversational agents and affective virtual agents, and affect-adaptive human-machine interaction. Since such technologies and their application to mental health also gives rise to important ethical concerns, this workshop aims to identify and address these emerging issues.

Workshop Website:

Workshop Chair: Iulia Lefter

What’s Next in Affect Modeling?


The valid and reliable evaluation of affect and affective interaction is key for the advancement of affective computing (AC). Recent breakthroughs in deep (machine) learning and generative AI have boosted the efficiency and generality of affect models by discovering novel representations of users and their context acting on high resolutions of multimodal signals. Such representations, however, are data-hungry and in need of large datasets that AC is not able to offer. Moreover, as affect models gradually become larger and more complex, their expressivity, explainability, and transparency become increasingly opaque. This workshop series puts an emphasis on state-of-the-art methods in machine learning and their suitability for advancing the reliability, validity, and generality of affective models. We will be investigating entirely new methods, untried in AC, but also methods that can be coupled with traditional and dominant practices in affective modeling. In particular, we encourage submissions that offer visions of particular algorithmic advancements for affect modeling and proof-of-concept case studies showcasing the potential of new sophisticated machine learning methods. This is the third workshop in the series after the successful first and second events organised in conjunction with ACII 2021 and 2022. Papers were submitted, and after a double-blind peer-review process (three reviewers assigned to each paper), the accepted ones were presented during the event. Besides paper presentations, the workshop featured a keynote talk delivered by Prof. Michel Valstar (2021) and Prof. Erik Cambria (2022). Around twenty-five participants attended each of the workshops.

Workshop Website:

Workshop Chair: Matthew Barthet

Moral Imagination in Affective Computing


The values and perspectives of affective technologists often shape what capabilities are built and how they are deployed. While most researchers want to see their work drive socially beneficial impacts, the technologies are being deployed in many complex social contexts. If our assumptions and intuitions go unchallenged, they can and do lead to unintended adverse effects. Realizing socially beneficial affective technologies requires going beyond dominant norms of computer science and engineering culture to engage meaningfully and productively with ethics and social responsibility. Using a “Moral Imagination” methodology developed at Google, this workshop provides an interactive format in which the community can engage pragmatically and constructively with this critical part of affective computing research.

Workshop Website:

Workshop Chair: Amanda McCroskery