Overview

Learning occurs throughout our entire life. We all learn about different things, at our own pace and method. In classical education, a class of learners is meant to learn a standard course in a pace consistent for the entire class. However, when a group of students are taught the same lessons at the same pace some students may lag in comprehension while some others excel. The concept of adaptive learning originates from the fact that every learner has a unique learning curve and personalized content aligned to her learning curve can improve learning outcomes. Recent technologies have made it possible to capture learner’s data, on the basis of which the system delivers different topics to the learner.

The Robots for Learning workshop, in its 5th series, focuses on adaptive learning. Adaptive learning systems are envisioned to maximize learning efficiency, improve learning outcomes, create personalised learning experience, generate analytics for early interventions, increase engagement, and allow for customizability to fit individual needs and preferences. In this workshop, we aim to discuss the approaches and challenges of using robots for learner-centric methods, learning path adaptation, social personalisation, remediation and multimodal techniques. Robots can be able to perceive and monitor learning, and at the same time empathize and give personalized feedback to the learners.

With this workshop, we aim at discussing recent advances in empirical and theoretical state-of-the-art research contributions on HRI in educational contexts regarding the following challenges: how adaptivity can play a role in the educational context? How can we use machine learning to provide better learning? How could robots be used to foster adaptive learning for a group of users?

List of topics

• Adaptive mechanisms for robot tutors, personalization and adaptation algorithms for tutoring interactions
• Design of autonomous systems for tutoring interactions
• Theories and methods for tutoring (pedagogical and language acquisition)
• Shared knowledge and knowledge modelling in HRI
• Human-robot collaborative learning
• Attachment and learning with a social robot (social and cognitive development)
• Engagement in educational human-robot interaction
• Human-robot relationship assessment
• Designing student models and assessing student’s learning
• Playful learning with a robot
• Human-robot creativity
• Kinesthetic and non-verbal communication in human-robot interaction
• Impact of embodiment on learning
• Technical innovation in learning or teaching robots
• Long term learning interactions, design and methodologies for repeated human-robot encounters
• Robots for learners with special needs and special abilities
• Education and re-training for adults
• Rehabilitation and re-education
• Privacy and ethical issues in robot tutoring applications

Program

9:00-9:05 Introduction 9:05-9:40 Keynote Tony Belpaeme (30 min +5) 9:40-10:30 Presentations (2 papers, 20 min +5 each) 10:30-11:00 Coffee break 11:00-11:50 Presentations (2 papers, 20 min +5 each) 11:50-12:50 Semi-structured group discussions 12:50-13:00 Conclusion & Wrap up

Selected Papers

1.  Children’s anthropomorphism of second language robot tutors (Mirjam de Haas, Rianne van den Berghe, Paul Leseman, Emiel Krahmer, Josje Verhagen, Paul Vogt, Bram Willemsen and Jan de Wit)
   

2.  Storytelling with a social robot (Nicole Goossens, Rian Aarts and Paul Vogt) 
   

3.  Effective robot-human skill transfer via mutual reinforcement learning (Sayanti Roy, Emily Kieson, Charles Abramson and Christopher Crick)