Technology Category
Data Mining & Knowledge Management: Anomaly/Outlier DetectionComputer Vision: Generative Adversarial Networks (GANs) for VisionMachine Learning: Deep Generative Models & Autoencoders
ARES: Auxiliary Range Expansion for Outlier Synthesis
🤖 AI Summary
To address the degraded out-of-distribution (OOD) detection performance in open-set scenarios where in-distribution (ID) training data are unavailable, this paper proposes a distribution-free virtual anomaly generation method. The approach introduces three key innovations: (1) a latent-space “range-escape” strategy that actively steers anomaly synthesis away from ID regions—thereby modeling an OOD manifold beyond conventional boundary-neighborhood synthesis; (2) an energy-score-based discriminator to enhance robust separation between ID and OOD samples; and (3) a multi-stage anomaly generation framework incorporating, for the first time, an auxiliary range-expansion mechanism to broaden coverage of anomalous semantics. Extensive experiments on multiple benchmark datasets demonstrate significant improvements in OOD detection accuracy. Qualitative analysis further validates the plausibility and interpretability of the generated anomalies.
Technology Category
Application Category
📝 Abstract
Recent successes of artificial intelligence and deep learning often depend on the well-collected training dataset which is assumed to have an identical distribution with the test dataset. However, this assumption, which is called closed-set learning, is hard to meet in realistic scenarios for deploying deep learning models. As one of the solutions to mitigate this assumption, research on out-of-distribution (OOD) detection has been actively explored in various domains. In OOD detection, we assume that we are given the data of a new class that was not seen in the training phase, i.e., outlier, at the evaluation phase. The ultimate goal of OOD detection is to detect and classify such unseen outlier data as a novel"unknown"class. Among various research branches for OOD detection, generating a virtual outlier during the training phase has been proposed. However, conventional generation-based methodologies utilize in-distribution training dataset to imitate outlier instances, which limits the quality of the synthesized virtual outlier instance itself. In this paper, we propose a novel methodology for OOD detection named Auxiliary Range Expansion for Outlier Synthesis, or ARES. ARES models the region for generating out-of-distribution instances by escaping from the given in-distribution region; instead of remaining near the boundary of in-distribution region. Various stages consists ARES to ultimately generate valuable OOD-like virtual instances. The energy score-based discriminator is then trained to effectively separate in-distribution data and outlier data. Quantitative experiments on broad settings show the improvement of performance by our method, and qualitative results provide logical explanations of the mechanism behind it.
Problem
Research questions and friction points this paper is trying to address.
Deep Learning
Anomaly Detection
Data Distribution Mismatch
Innovation
Methods, ideas, or system contributions that make the work stand out.
ARES
Out-of-Distribution Detection
Virtual Outlier Synthesis
🔎 Similar Papers
No similar papers found.
