Isolation forest (iForest) has been emerging as arguably the most popular anomaly detector in recent years due to its general effectiveness across different benchmarks and strong scalability. Nevertheless, its linear axis-parallel isolation method often leads to (i) failure in detecting hard anomalies that are difficult to isolate in high-dimensional/non-linear-separable data space, and (ii) notorious algorithmic bias that assigns unexpectedly lower anomaly scores to artefact regions. These issues contribute to high false negative errors. Several iForest extensions are introduced, but they essentially still employ shallow, linear data partition, restricting their power in isolating true anomalies. Therefore, this paper proposes deep isolation forest. We introduce a new representation scheme that utilises casually initialised neural networks to map original data into random representation ensembles, where random axis-parallel cuts are subsequently applied to perform the data partition. This representation scheme facilitates high freedom of the partition in the original data space (equivalent to non-linear partition on subspaces of varying sizes), encouraging a unique synergy between random representations and random partition-based isolation. Extensive experiments show that our model achieves significant improvement over state-of-the-art isolation-based methods and deep detectors on tabular, graph and time series datasets; our model also inherits desired scalability from iForest.

Time series anomaly detection is instrumental in maintaining system availability in various domains. Current work in this research line mainly focuses on learning data normality deeply and comprehensively by devising advanced neural network structures and new reconstruction/prediction learning objectives. However, their one-class learning process can be misled by latent anomalies in training data (i.e., anomaly contamination) under the unsupervised paradigm. Their learning process also lacks knowledge about the anomalies. Consequently, they often learn a biased, inaccurate normality boundary. To tackle these problems, this paper proposes calibrated one-class classification for anomaly detection, realizing contamination-tolerant, anomaly-informed learning of data normality via uncertainty modeling-based calibration and native anomaly-based calibration. Specifically, our approach adaptively penalizes uncertain predictions to restrain irregular samples in anomaly contamination during optimization, while simultaneously encouraging confident predictions on regular samples to ensure effective normality learning. This largely alleviates the negative impact of anomaly contamination. Our approach also creates native anomaly examples via perturbation to simulate time series abnormal behaviors. Through discriminating these dummy anomalies, our one-class learning is further calibrated to form a more precise normality boundary. Extensive experiments on ten real-world datasets show that our model achieves substantial improvement over sixteen state-of-the-art contenders.

Background: One of the most cost-effective treatments for secondary prevention of stroke and other non-communicable diseases is a long-term medication regimen. However, the complexities of medication adherence extend far beyond individual behavior change, particularly in low- and middle-income countries. The purpose of this study was to examine stakeholder perspectives on barriers to medication adherence for stroke patients in Beijing, China, identifying opportunities to improve care and policy in resource-constrained settings.Methods: We conducted a qualitative, phenomenological analysis of data obtained from 36 individuals. Participants were patients; caregivers; healthcare providers; and representatives from industry and government, purposively selected to synthesize multiple perspectives on medication management and adherence for stroke secondary prevention in Beijing, China. Data was analyzed by thematic analysis across iterative coding cycles.Results: Four major themes characterized barriers on medication adherence, across stakeholders and geographies: limitations driven by individual patient knowledge / attitudes; lack of patient-provider interaction time; lack of coordination across the stratified health system; and lack of affordability driven by high overall costs and limited insurance policies.Conclusions: These barriers to medication management and adherence suggest opportunities for policy reform and local practice changes, particularly for multi-tiered health systems. Findings from this study in Beijing, China could be explored for applicability in other low- and middle-income countries with urban centers serving large geographic regions.