TY - GEN
T1 - Rethinking the Evaluation of Non-stationary Dueling Bandits for Human-Robot Interaction
AU - Schneider, Sebastian
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.
PY - 2026/1/2
Y1 - 2026/1/2
N2 - Non-stationary dueling bandits are promising for modeling evolving preferences in human-robot interaction, but current evaluations rely on synthetic data and ignore user experience. This short contribution proposes a regret formulation that incorporates interaction costs—such as annoyance, cognitive effort, and trust degradation. Furthermore, suggestions for standardized, user-in-the-loop evaluation is presented. Recognizing the challenges of long-term studies with real users, large language models could be used to simulate non-stationary user behavior for early-stage testing. This enables deeper insight into algorithm usability and supports the development of adaptive systems that are not only efficient, but also aligned with human expectations and experience.
AB - Non-stationary dueling bandits are promising for modeling evolving preferences in human-robot interaction, but current evaluations rely on synthetic data and ignore user experience. This short contribution proposes a regret formulation that incorporates interaction costs—such as annoyance, cognitive effort, and trust degradation. Furthermore, suggestions for standardized, user-in-the-loop evaluation is presented. Recognizing the challenges of long-term studies with real users, large language models could be used to simulate non-stationary user behavior for early-stage testing. This enables deeper insight into algorithm usability and supports the development of adaptive systems that are not only efficient, but also aligned with human expectations and experience.
KW - 2026 OA procedure
KW - Human-Robot Interaction
KW - Preference Learning
KW - Evaluation Methods
UR - https://www.scopus.com/pages/publications/105026338799
U2 - 10.1007/978-981-95-2379-5_31
DO - 10.1007/978-981-95-2379-5_31
M3 - Conference contribution
AN - SCOPUS:105026338799
SN - 978-981-95-2378-8
T3 - Lecture Notes in Computer Science
SP - 450
EP - 455
BT - Social Robotics + AI
A2 - Staffa, Mariacarla
A2 - Cabibihan, John-John
A2 - Siciliano, Bruno
A2 - Ge, Shuzhi Sam
A2 - Bodenhagen, Leon
A2 - Tapus, Adriana
A2 - Rossi, Silvia
A2 - Cavallo, Filippo
A2 - Fiorini, Laura
A2 - Matarese, Marco
A2 - He, Hongsheng
PB - Springer
CY - Singapore
T2 - 17th International Conference on Social Robotics, ICSR+AI 2025
Y2 - 10 September 2025 through 12 September 2025
ER -