Unseen Class Discovery in Open-world Classification

By Lei Shu, Hu Xu, B. Liu. 2018

76 citations recorded by the source corpus.

This paper concerns open-world classification, where the classifier not only needs to classify test examples into seen classes that have appeared in training but also reject examples from unseen or novel classes that have not appeared in training. Specifically, this paper focuses on discovering the hidden unseen classes of the rejected examples. Clearly, without prior knowledge this is difficult. However, we do have the data from the seen training classes, which can tell us what kind of similarity/difference is expected for examples from the same class or from different classes. It is reasonable to assume that this knowledge can be transferred to the rejected examples and used to discover the hidden unseen classes in them. This paper aims to solve this problem. It first proposes a joint open classification model with a sub-model for classifying whether a pair of examples belongs to the same or different classes. This sub-model can serve as a distance function for clustering to discover the hidden classes of the rejected examples. Experimental results show that the proposed model is highly promising.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Bing Liu
2. Lei Shu
3. Sahisnu Mazumder
4. Zixuan Ke
5. Eric Robertson
6. Hu Xu
7. Qi Qin
8. Scott Grigsby

Citations in the corpus

• Learngene: From Open-World to Your Learning Task (2021, background): “open world recognition or classification) [4, 42, 55] can be broadly defined as learning a model that can perform its intended task and then incrementally learn the new things.”
• Learning on the Job: Online Lifelong and Continual Learning (2020, background): “Steps 1 and 3 have been studied in open-world learning (Shu et al., 2017a, 2018; Chen and Liu, 2018) and continual learning (Chen and Liu, 2018; Parisi et al., 2018) respectively.”
• Recent Advances in Open Set Recognition: A Survey (2018, methodology): “To address these problems, researchers have made many efforts [58], [71]–[77], [97].”
• Classification-Reconstruction Learning for Open-Set Recognition (2018, methodology): “DOC (deep open classifier) [40, 41], which is designed for document classification, enables end-to-end training by eliminating outlier detectors outside networks and using sigmoid activations in the networks for performing joint classification and outlier detection.”
• Self-Initiated Open World Learning for Autonomous AI Agents (2021, background): “Some researchers have further studied how to automatically identify the classes or categories of the detected novel object instances (Shu, Xu, and Liu 2018).”
• HRN: A Holistic Approach to One Class Learning (2020, methodology): “Some even use anomalies in the training data to build multi-class classifiers [76, 36, 62].”
• Zero-Shot Out-of-Distribution Detection Based on the Pre-trained Model CLIP (2021, methodology): “Recent methods are mainly based on deep learning (Shu, Xu, and Liu 2017, 2018; Xu et al. 2019; Shu et al. 2021; Liang, Li, and Srikant 2017; Per-era et al. 2020; Miller et al. 2021) and try to solve the problem from different perspectives.”