Co-Adaptive Eco-Nudging: A Privacy-Preserving Contextual Bandit with User-Taught Preferences (CCF A)
Published in CHI Conference on Human Factors in Computing Systems 2026, 2026
Two field studies with an on-device contextual bandit for personalized eco-nudging. This paper introduces the Ethical-Efficacy Frontier (EEF) and Energy ROI accounting for sustainable behavior change interventions.
Keywords: Behavior Change, Personalization, Sustainability, Nudging, Privacy-Preserving AI, Contextual Bandits
Subcommittee: Critical Computing, Sustainability, & Social Justice; Understanding People: Mixed & Alternative Methods
Recommended citation: Fan, G., Liu, D., et al. (2026). Co-Adaptive Eco-Nudging: A Privacy-Preserving Contextual Bandit with User-Taught Preferences. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI 2026). ACM.