On-device training enables the model to adapt to user-specific data by fine-tuning a pre-trained model locally. As embedded devices become ubiquitous, on-device training is increasingly essential since users can benefit from the personalized model without transmitting data and model parameters to the server. Despite significant efforts toward efficient training, ondevice training still faces a major challenge: The prohibitive cost of multi-layer backpropagation strains the limited resources of tiny devices. In this paper, we propose an algorithm-system cooptimization framework TinyMP that enables self-adaptive on-tiny-device model personalization. To mitigate backpropagation costs, we introduce Gradient Condensing to condense the gradient map structure, significantly reducing the computational complexity and memory consumption of backpropagation while preserving model performance. To further reduce computation overhead, we propose Alternant Partial Update, a mechanism that locally and alternatively selects essential parameters to update without requiring retraining or offline evolutionary search. Our framework is evaluated through extensive experiments using various CNN models (e.g., MobileNetV2, MCUNet) on embedded devices with minimal resources (e.g., OpenMV-H7 with less than 1MB SRAM and 2 MB Flash). Experimental results show that our framework achieves up to $2.4 \times$ speedup, 80.8% memory saving, and 30.3% accuracy improvement on downstream tasks, outperforming SOTA approaches.

LINE has garnered over 700 million users globally. Research on LINE user geolocation can not only evaluate the effectiveness of privacy protection strategies but also support governmental efforts in cybercrime investigations and forensics. Existing instant messaging user geolocation methods typically rely on a consistent correlation between the reported and actual distances in features such as the "Nearby People" service. However, these methods struggle to handle LINE’s location obfuscation characteristics, such as large variations in reported distance granularity, weak regularity in user ordering, and random user appearance probabilities, making them unsuitable for directly locating LINE users. To address these challenges, this paper introduces a LINE user geolocation method based on Probe Position Adaptive Adjustment (P2A2). Through the analysis of the historical state information, including historical movement directions of probe and historical reported distances of target, during the geolocation process, P2A2 determines the optimal method for the probe’s movement direction and dynamically adjusts its direction and step length. This method accelerates the probe’s convergence to the target’s position, significantly reduces inefficient movements, and enhances geolocation efficiency while maintaining high accuracy. The experimental evaluation conducted on the LINE application shows that P2A2 achieves high efficiency and precision in target geolocation, with an average geolocation error of 83.8 meters. Compared to representative methods such as MSAG, RRABG, and HNBG, the geolocation efficiency of P2A2 is improved by 39.3%-86.7%, and the average geolocation error is reduced by 43.5%-64.9%.

Adverse drug reactions (ADRs) are among the global public health events that seriously endanger human life and cause high economic burdens. Therefore, predicting the possibility of their occurrence and taking early and effective response measures is of great significance. Constructing a correlation matrix between drugs and their adverse reactions, followed by effective correlation data mining, is one of the current strategies to predict ADRs using accessible public data. Since the number of known ADRs in real-world data is far less than the number of their unknown counterparts, the drug-ADR association matrix is very sparse, which greatly affects the classification performance of machine learning methods. To effectively address the problem of sparsity, we proposed a novel weighted pseudo-labeling framework that mines potential unknown drug-ADR pairs by integrating multiple weighted matrix factorization (MF) models and treating them as pseudo-labeled drug-ADR pairs. Pseudo-labeled data is added to the training set, and the MF model is fine-tuned to improve the classification performance. To prevent overfitting to easily found pseudo-labels and improve the quality of pseudo-labels, a novel weighting approach for pseudo-labels was adopted. This paper reproduces the baselines under the same experimental conditions to evaluate the performance of the proposed method on sparse data from the Side Effect Resource (SIDER) database. Experimental results showed that our method outperformed other baselines in the Area Under Precision-Recall and F1-scores and still maintained the best performance in sparser scenarios. Furthermore, we conducted a case study, and the results showed that our proposed framework efficiently predicted ADRs in the real world.

Biomedical texts provide important data for investigating drug-drug interactions (DDIs) in the field of pharmacovigilance. Although researchers have attempted to investigate DDIs from biomedical texts and predict unknown DDIs, the lack of accurate manual annotations significantly hinders the performance of machine learning algorithms. In this study, a new DDI prediction framework, Subgraph Enhance model, was developed for DDI (SubGE-DDI) to improve the performance of machine learning algorithms. This model uses drug pairs knowledge subgraph information to achieve large-scale plain text prediction without many annotations. This model treats DDI prediction as a multi-class classification problem and predicts the specific DDI type for each drug pair (e.g. Mechanism, Effect, Advise, Interact and Negative). The drug pairs knowledge subgraph was derived from a huge drug knowledge graph containing various public datasets, such as DrugBank, TwoSIDES, OffSIDES, DrugCentral, EntrezeGene, SMPDB (The Small Molecule Pathway Database), CTD (The Comparative Toxicogenomics Database) and SIDER. The SubGE-DDI was evaluated from the public dataset (SemEval-2013 Task 9 dataset) and then compared with other state-of-the-art baselines. SubGE-DDI achieves 83.91% micro F1 score and 84.75% macro F1 score in the test dataset, outperforming the other state-of-the-art baselines. These findings show that the proposed drug pairs knowledge subgraph-assisted model can effectively improve the prediction performance of DDIs from biomedical texts.