🤖 AI Summary
This study addresses the limitations of existing outlier generation methods, which often rely on heuristics, exhibit instability, and struggle to cover extreme out-of-distribution scenarios. To overcome these challenges, this work proposes the SBOG framework, which pioneers the use of Sinkhorn optimal transport-induced support costs to guide sampling toward weakly supported boundary regions. By integrating distributionally robust optimization with latent-space semantic constraints, SBOG generates controlled, structured outliers in the latent space, thereby transcending the limitations of conventional arbitrary sparse sampling. Experimental results demonstrate that SBOG produces informative and semantically controllable samples in both time-series anomaly and image outlier synthesis tasks, significantly enhancing the capability for robustness evaluation across downstream cross-modal applications.
📝 Abstract
Outliers are essential for evaluating and improving the robustness of machine learning systems, especially when future distributions may differ significantly from historical training data. In high-stakes applications, robustness often depends on rare cases that finite datasets fail to capture, making simple resampling or perturbation insufficient for stress scenario generation. Existing outlier synthesis methods typically rely on sparse neighborhoods, low support latent regions, or classifier boundary crossings, which can be heuristic, unstable, and tied to specific modalities or architectures. We therefore propose Sinkhorn Boundary Outlier Generation (SBOG), a structured framework for latent-space outlier generation that couples Sinkhorn optimal transport geometry with distributionally robust boundary modeling. The resulting Sinkhorn-induced support cost guides the sampler toward weakly supported boundary regions, while semantic constraints prevent uncontrolled drift from the intended context, yielding controlled deviations from the in-distribution reference measure rather than arbitrary sparse-region samples. Experiments on time series anomaly generation and image outlier synthesis show that our framework produces informative, semantically controlled outliers and improves downstream robustness evaluation across modalities, providing a foundation for stress scenario generation beyond empirical support.