Robust Data-Driven Automation Based on Relaxed Supervised Hashing With Self-Optimized Labels

• Published in: IEEE transactions on automation science and engineering 2024-01, Vol.21 (1), p.512-527
• Main Authors: Chen, Bo-Wei, Huang, Jhao-Yang, Wu, Ying-Hsuan
• Format: Article
• Language: English
• Subjects: Actuators; Automation; Codes; Complexity theory; half quadratic optimization; incomplete data; Internet of Things; Labels; Learning; Loss measurement; Modelling; Occlusion; Outliers (statistics); partially observed data; Quadratic programming; relaxation; Relaxation methods; relaxed robust supervised hashing; Robust data-driven automation; Robustness; Signatures; Slack variables; Supervised learning; Visual Internet of Things; Visualization
• ISSN: 1545-5955; 1558-3783
• DOI: 10.1109/TASE.2022.3221311

Recently, the Visual Internet of Things (VIoT) has been widely used in data-driven automation, where VIoT devices are used to monitor environmental dynamics and to trigger corresponding actuators after examining event signatures (e.g., hashes). Nonetheless, VIoT devices may collect partially observed data, e.g., largely occluded images, which cause biased labeling and oversensitivity during modeling. Moreover, in typical methods, class labels are rigid and fixed categorical variables, where marginal space between different classes is fixed and equal. In fact, margins may be unequally spaced. Such nonflexibility in class labels inevitably causes fitting difficulty. In light of such, this study proposes relaxed robust supervised hashing (Relaxed RSH) for generating reliable signatures that can simultaneously conquer the above problems caused by incomplete data and rigid margins. To accommodate oversensitivity, this study proposes measuring hash learning loss by robust half quadratic (HQ) functions for modeling incomplete data. In the initial label matrix, slack variables are added to relax binary constraints. Such slack variables can be self-optimized during the learning process and can be used to automatically adjust margins between different classes. Decorrelation, balancing, and normalization constraints based on Relaxed RSH are also devised to provide discriminant and compact codes. Experimental results based on open datasets showed that the proposed method yielded higher mAP and F1 than the baselines. Note to Practitioners-This work is motivated by the problems caused by incomplete data and rigid class margins during event signature (e.g., hash) learning, where hashes are used to trigger automation systems. In existing hashing methods, incomplete data (e.g., continuous occlusion, missing values, and sample-specific outliers) along with rigid class margins cause fitting biases, thereby challenging data-driven automation. This study proposes Relaxed RSH by designing robust HQ loss, self-optimized label learning, and corresponding constraints for generating discriminant hashes. This prevents actuators from being mistriggered by corrupted data. Experiments were conducted on various data corruption. Future research will address the design for incremental/decentralized hash learning.