Search for: All records

Membership inference attacks (MIAs) are currently considered one of the main privacy attack strategies, and their defense mechanisms have also been extensively explored. However, there is still a gap between the existing defense approaches and ideal models in both performance and deployment costs. In particular, we observed that the privacy vulnerability of the model is closely correlated with the gap between the model's data-memorizing ability and generalization ability. To address it, we propose a new architecture-agnostic training paradigm called Center-based Relaxed Learning (CRL), which is adaptive to any classification model and provides privacy preservation by sacrificing a minimal or no loss of model generalizability. We emphasize that CRL can better maintain the model's consistency between member and non-member data. Through extensive experiments on common classification datasets, we empirically show that this approach exhibits comparable performance without requiring additional model capacity or data costs. 

Augmented Reality (AR) technology offers the possibility of experiencing virtual images with physical objects and provides high quality hands-on experiences in an engineering lab environment. However, students still need help navigating the educational content in AR environments due to a mismatch problem between computer-generated 3D images and actual physical objects. This limitation could significantly influence their learning processes and workload in AR learning. In addition, a lack of student awareness of their learning process in AR environments could negatively impact their performance improvement. To overcome those challenges, we introduced a virtual instructor in each AR module and asked a metacognitive question to improve students’ metacognitive skills. The results showed that student workload was significantly reduced when a virtual instructor guided students during AR learning. Also, there is a significant correlation between student learning performance and workload when they are overconfident. The outcome of this study will provide knowledge to improve the AR learning environment in higher education settings. 

This study aims to develop an interactive learning solution for engineering education by combining augmented reality (AR), Near-Field Electromagnetic Ranging (NFER), and motion capture technologies. We built an instructional system that integrates AR devices and real-time positioning sensors to improve the interactive experience of learners in an immersive learning environment, while the motion, eye-tracking, and location-tracking data collected by the devices applied to learners enable instructors to understand their learning patterns. To test the usability of the system, two AR-based lectures were developed with different difficulty levels (Lecture 1 - Easy vs. Lecture 2 - Hard), and the System Usability Scale (SUS) was collected from thirty participants. We did not observe a significant usability difference between Lecture 1 and Lecture 2. Through the experiment, we demonstrated the robustness of this AR learning system and its unique promise in integrating AR teaching with other technologies. 

Being conscious of your thought processes is known as metacognition. It supports students in being more aware of their actions, motivations, and the potential applications of the skills [1]. This study investigates how different metacognitive judgment questions affect students’ metacognitive awareness in an augmented reality (AR) environment. The outcomes of this study will help us to understand what metacognitive monitoring method is more effective in the AR learning environment. According to the literature, students with high knowledge about cognition have higher test performance, while students with low regulation have a challenge during planning, organizing, and elaborating strategies. The dependent variables of the study are student learning performance and metacognitive awareness inventory (MAI) score, and one independent variable is the metacognitive judgment question Retrospective Confidence Judgment (RCJ) and Judgment of Learning (JOL). We hypothesized that the students with high performance would have improved MAI scores in both groups. The experiment was done with two groups (RCJ and JOL). Both groups responded to the pre-post metacognitive awareness inventory questionnaire. During the experiment, the MAI questionnaire was asked two times. In round one, the MAI questionnaire was asked at the beginning of lecture one; however, in round two, the questionnaire was asked at the end of lecture two. Results indicated significant differences in RCJ low performers. In RCJ, the participants whose performance was significantly reduced in lecture 2 had a higher improvement on MAI both regulation and knowledge about cognition. Overall, the result of our study could advance our understanding of how to design an advanced instructional strategy in an AR environment.