Search papers, labs, and topics across Lattice.
This paper introduces a dual-stage energy-based active learning (AL) framework to address the challenge of open-set conditions, where unlabeled data contains both known and unknown classes. The framework uses two energy-based models (EBMs): one to filter out unknown class samples and another to score the informativeness of known samples based on their energy landscape. Experiments on image classification datasets (CIFAR-10, CIFAR-100, TinyImageNet, ModelNet40) demonstrate that the proposed framework achieves superior annotation efficiency and classification performance compared to existing open-set AL methods.
Querying strategically with energy-based models lets you actively learn in the wild, even when you don't know what you don't know.
Active learning (AL) has emerged as a crucial methodology for minimizing labeling costs in deep learning by selecting the most valuable samples from a pool of unlabeled data for annotation. Traditional AL operates under a closed-set assumption, where all classes in the dataset are known and consistent. However, real-world scenarios often present open-set conditions in which unlabeled data contains both known and unknown classes. In such environments, standard AL techniques struggle. They can mistakenly query samples from unknown categories, leading to inefficient use of annotation budgets. In this paper, we propose a novel dual-stage energy-based framework for open-set AL. Our method employs two specialized energy-based models (EBMs). The first, an energy-based known/unknown separator, filters out samples likely to belong to unknown classes. The second, an energy-based sample scorer, assesses the informativeness of the filtered known samples. Using the energy landscape, our models distinguish between data points from known and unknown classes in the unlabeled pool by assigning lower energy to known samples and higher energy to unknown samples, ensuring that only samples from classes of interest are selected for labeling. By integrating these components, our approach ensures efficient and targeted sample selection, maximizing learning impact in each iteration. Experiments on 2D (CIFAR-10, CIFAR-100, TinyImageNet) and 3D (ModelNet40) object classification benchmarks demonstrates that our framework outperforms existing approaches, achieving superior annotation efficiency and classification performance in open-set environments.
