Feature Decoupling in Self-supervised Representation Learning for Open Set Recognition

• URL: http://arxiv.org/abs/2209.14385v1
• Date: Wed, 28 Sep 2022 19:21:53 GMT
• Title: Feature Decoupling in Self-supervised Representation Learning for Open Set Recognition
• Authors: Jingyun Jia, Philip K. Chan
• Abstract summary: We use a two-stage training strategy for the open set recognition (OSR) problems. In the first stage, we introduce a self-supervised feature decoupling method that finds the content features of the input samples from the known classes. In the second stage, we fine-tune the content features with the class labels. Our experimental results indicate that our proposed self-supervised approach outperforms others in image and malware OSR problems.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Assuming unknown classes could be present during classification, the open set recognition (OSR) task aims to classify an instance into a known class or reject it as unknown. In this paper, we use a two-stage training strategy for the OSR problems. In the first stage, we introduce a self-supervised feature decoupling method that finds the content features of the input samples from the known classes. Specifically, our feature decoupling approach learns a representation that can be split into content features and transformation features. In the second stage, we fine-tune the content features with the class labels. The fine-tuned content features are then used for the OSR problems. Moreover, we consider an unsupervised OSR scenario, where we cluster the content features learned from the first stage. To measure representation quality, we introduce intra-inter ratio (IIR). Our experimental results indicate that our proposed self-supervised approach outperforms others in image and malware OSR problems. Also, our analyses indicate that IIR is correlated with OSR performance.

Related papers

• CREST: Cross-modal Resonance through Evidential Deep Learning for Enhanced Zero-Shot Learning [48.46511584490582]
  Zero-shot learning (ZSL) enables the recognition of novel classes by leveraging semantic knowledge transfer from known to unknown categories. Real-world challenges such as distribution imbalances and attribute co-occurrence hinder the discernment of local variances in images. We propose a bidirectional cross-modal ZSL approach CREST to overcome these challenges.
  arXiv  Detail & Related papers  (2024-04-15T10:19:39Z)
• Learning Adversarial Semantic Embeddings for Zero-Shot Recognition in Open Worlds [25.132219723741024]
  Zero-Shot Learning (ZSL) focuses on classifying samples of unseen classes with only their side semantic information presented during training. "Zero-Shot Open-Set Recognition" (ZS-OSR) is required to accurately classify samples from the unseen classes while rejecting samples from the unknown classes during inference. We introduce a novel approach specifically designed for ZS-OSR, in which our model learns to generate adversarial semantic embeddings of the unknown classes to train an unknowns-informed ZS-OSR.
  arXiv  Detail & Related papers  (2023-07-07T06:54:21Z)
• Enlarging Instance-specific and Class-specific Information for Open-set Action Recognition [47.69171542776917]
  We find that features with richer semantic diversity can significantly improve the open-set performance under the same uncertainty scores. A novel Prototypical Similarity Learning (PSL) framework is proposed to keep the instance variance within the same class to retain more IS information.
  arXiv  Detail & Related papers  (2023-03-25T04:07:36Z)
• OvarNet: Towards Open-vocabulary Object Attribute Recognition [42.90477523238336]
  We propose a naive two-stage approach for open-vocabulary object detection and attribute classification, termed CLIP-Attr. The candidate objects are first proposed with an offline RPN and later classified for semantic category and attributes. We show that recognition of semantic category and attributes is complementary for visual scene understanding.
  arXiv  Detail & Related papers  (2023-01-23T15:59:29Z)
• Open Set Recognition using Vision Transformer with an Additional Detection Head [6.476341388938684]
  We propose a novel approach to open set recognition (OSR) based on the vision transformer (ViT) technique. Our approach employs two separate training stages. First, a ViT model is trained to perform closed set classification. Then, an additional detection head is attached to the embedded features extracted by the ViT, trained to force the representations of known data to class-specific clusters compactly.
  arXiv  Detail & Related papers  (2022-03-16T07:34:58Z)
• M2IOSR: Maximal Mutual Information Open Set Recognition [47.1393314282815]
  We propose a mutual information-based method with a streamlined architecture for open set recognition. The proposed method significantly improves the performance of baselines and achieves new state-of-the-art results on several benchmarks consistently.
  arXiv  Detail & Related papers  (2021-08-05T05:08:12Z)
• SCARF: Self-Supervised Contrastive Learning using Random Feature Corruption [72.35532598131176]
  We propose SCARF, a technique for contrastive learning, where views are formed by corrupting a random subset of features. We show that SCARF complements existing strategies and outperforms alternatives like autoencoders.
  arXiv  Detail & Related papers  (2021-06-29T08:08:33Z)
• Neighborhood Contrastive Learning for Novel Class Discovery [79.14767688903028]
  We build a new framework, named Neighborhood Contrastive Learning, to learn discriminative representations that are important to clustering performance. We experimentally demonstrate that these two ingredients significantly contribute to clustering performance and lead our model to outperform state-of-the-art methods by a large margin.
  arXiv  Detail & Related papers  (2021-06-20T17:34:55Z)
• SCAN: Learning to Classify Images without Labels [73.69513783788622]
  We advocate a two-step approach where feature learning and clustering are decoupled. A self-supervised task from representation learning is employed to obtain semantically meaningful features. We obtain promising results on ImageNet, and outperform several semi-supervised learning methods in the low-data regime.
  arXiv  Detail & Related papers  (2020-05-25T18:12:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.