VPL System Personalizes Vibrotactile Feedback via Preference Learning

Researchers have developed Vibrotactile Preference Learning (VPL), a system that personalizes vibration feedback using Gaussian-process-based uncertainty-aware preference learning. VPL captures user-specific preferences over vibrotactile parameters through an expected information gain-based acquisition strategy, guiding query selection over 40 rounds of pairwise comparisons. The system incorporates user-reported uncertainty to efficiently explore the parameter space. In a user study with 13 participants using a Microsoft Xbox controller, VPL demonstrated effective learning of individualized preferences while maintaining comfortable, low-workload interactions. The results suggest VPL's potential for scalable personalization of vibrotactile experiences in interactive systems.