Embedded Files
KIM BARAKA
VRIJE UNIVERSITEIT AMSTERDAMKIM BARAKA
Kim Baraka is a tenured assistant professor in the Computer Science Department at VU Amsterdam. He was a postdoctoral fellow in the Socially Intelligent Machines Lab at UT Austin, and holds a dual Ph.D. in Robotics from Carnegie Mellon University (CMU) and the Instituto Superior Técnico in Lisbon, Portugal. His interdisciplinary research, at the cross-roads of human-centered robotics, machine learning, and dance, focuses on enabling robots and humans to teach and learn from each other through situated social interactions. He is part of the Hybrid Intelligence Center (which fundamentally researches hybrid human-AI systems, with applications to healthcare and education), an Associate Editor of the ACM Transactions on Human-Robot Interaction, and a scientific board member of the Lorentz Center in Leiden. As a professionally trained contemporary dancer, he is particularly interested in new frontiers in robotics that draw inspiration from the performing arts.
Title of the Talk: Human-Interactive Robot Learning: Definition, Challenges, and Opportunities
Abstract: This talk will provide a broad overview of the emerging field of human-interactive robot learning (HIRL), an interdisciplinary effort aiming at building the next generation of robot “apprentices”. This talk will motivate the need for teachable embodied machines and outline challenges and opportunities for the field moving forward, with comparable attention to algorithms, interaction design, user modelling, and ethics and safety. This talk will be loosely based on a recent paper with the same title, and illustrated through research project from the speaker’s own research group.
Lola Canamero
Cergy Paris University, FranceLola Canamero
Lola Canamero is Full Professor and Chair of Neuroscience and Robotics at the Neurocybernetics Team of the ETIS Lab, CY Cergy Paris University, France, and (honorary) Visiting Professor at the University of Hertfordshire, UK, where she was faculty from 2001 to 2020. She holds a “Licenciatura” in Philosophy from the Complutense University of Madrid and a PhD in Computer Science from the University of Paris-XI. She turned to Embodied AI as a postdoct with Rodney Brooks at MIT AI-Lab (USA) and with Luc Steels at VUB AI-Lab (Belgium). Since 1995, her research investigates the interactions between motivation, emotion and embodied cognition from the perspectives of adaptation, development and evolution, using autonomous and social robots. Projects include the EU-funded HUMAINE (on emotion-oriented technology), FEELIX-GROWING (on emotion development in humans, non-human primates and robots), and ALIZ-E (on social companions for children with diabetes), or the UH-funded Autonomous Robots as Embodied Models of Mental Disorders. She has played a pioneering role in the emotion modeling community. She is author or co-author of over 150 peer-reviewed publications. Personal website: www.emotion-modeling.info/
Title of the Talk: You Are in My Body: Modeling Humans Without A User Model
Abstract: ...
DANIEL TOZADORE
University College, LondonDANIEL TOZADORE
Daniel Tozadore is Lecturer in Robotics & AI at UCL, with a PhD in Computer Science from the University of São Paulo (USP) and previous postdoctoral experience at EPFL. Over 10 years of experience in Computer Science, specialising in Robotic Systems, Human–Robot Interaction, AI for Education, deep learning, reinforcement learning, and adaptive systems. Has led major research projects including iReCHeCk (with Paris 8 University) on autonomous handwriting learning using social robots, and MI2US, investigating the role of social robots in supporting multicultural inclusion in Swiss schools. Was also founder and former CEO of TEIA Educacional, an AI-driven edtech startup recognised among Brazil’s top 10 most innovative education technologies in 2022, using adaptive and affective techniques in educational settings.
Title of the Talk: If Adaptation in Social Robots Were a Game, Who Would Be the Final Boss?
Abstract: Adaptive social robots are often presented as the next level of human–robot interaction: systems that can recognise behaviour, infer emotions and intentions, build user models, and personalise their actions over time. In education and socially assistive contexts, adaptation is usually framed as a path towards better engagement, personalised support, and more meaningful learning. But if adaptation in social robots were a game, who (or what) would be the final boss? This talk will use this metaphor to critically examine adaptive social robots through three perspectives. First, I will discuss what the literature tells us about the different “levels” of adaptation in HRI, from behaviour recognition and affective modelling to long-term personalisation, mutual adaptation, and ethically aware interaction. Second, I will reflect on my own experience designing and studying social robots and AI-based systems in educational contexts, including autonomous handwriting practice, personalised feedback, and robot-mediated inclusion in multicultural school environments. These examples will highlight the gap between the promise of adaptive systems and the practical, ethical, and pedagogical challenges of deploying them in real-world settings.
Finally, I will invite the audience to collectively reflect on what the final boss of adaptive social robotics might be: technical performance, user modelling, trust, context awareness, ethics, or perhaps knowing when not to adapt. I will argue for a shift from “more adaptation” to “better adaptation”: transparent, meaningful, inclusive, and human-centred forms of adaptation that support learning, trust, and human agency.
Google Sites
Report abuse