Select or Suggest? Reinforcement Learning-based Method for High-Accuracy Target Selection on Touchscreens
Year: 2022
Type: conference-paper
Abstract: Suggesting multiple target candidates based on touch input is a possible option for high-accuracy target selection on small touchscreen devices. But it can become overwhelming if suggestions are triggered too often. To address this, we propose SATS, a Suggestion-based Accurate Target Selection method, where target selection is formulated as a sequential decision problem. The objective is to maximize the utility: the negative time cost for the entire target selection procedure. The SATS decision process is dictated by a policy generated using reinforcement learning. It automatically decides when to provide suggestions and when to directly select the target. Our user studies show that SATS reduced error rate and selection time over Shift [51], a magnification-based method, and MUCS, a suggestion-based alternative that optimizes the utility for the current selection. SATS also significantly reduced error rate over BayesianCommand [58], which directly selects targets based on posteriors, with only a minor increase in selection time.
Authors: Zhi Li, Maozheng Zhao, Dibyendu Das, HANG ZHAO, Yan Ma, Wanyu Liu, Michel Beaudouin-Lafon, Fusheng Wang, IV Ramakrishnan, Xiaojun Bi
Institutions: Stony Brook University, Institut de Recherche et Coordination Acoustique Musique, Centre Pompidou, Sciences et Technologies de la Musique et du Son, Centre National de la Recherche Scientifique, Institut national de recherche en sciences et technologies du numérique, Université Paris-Saclay, Laboratoire Interdisciplinaire des Sciences du Numérique
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: Rethinking Electronic Fetal Monitoring to Improve Perinatal Outcomes and Reduce Frequency of Operative Vaginal and Cesarean Deliveries, Feel-IT: Audio-Tactile Computer Accessibility, Space-Compacting Magnification Augmented with Natural Gestures and Keyboardless Text Entry for Low Vision Smartphone Interaction, Bayesian-centric Multimodal Hands-free Computer Interaction Technologies for People with Quadriplegia, CHS: Small: Establishing Action Laws for Touch Interaction, Next Generation Screen Magnification Technology for People with Low Vision, SCC-PG: SafeGuard—Detecting and Minimizing Wandering Incidents of Children and Adults with Disabilities, G8850311024
OpenAlex ID: W4225112989
