Open-world Learning and Application to Product Classification

By Hu Xu, Bing Liu, Lei Shu, Philip S. Yu. 2018

119 citations recorded by the source corpus.

Classic supervised learning makes the closed-world assumption that the classes seen in testing must have appeared in training. However, this assumption is often violated in real-world applications. For example, in a social media site, new topics emerge constantly and in e-commerce, new categories of products appear daily. A model that cannot detect new/unseen topics or products is hard to function well in such open environments. A desirable model working in such environments must be able to (1) reject examples from unseen classes (not appeared in training) and (2) incrementally learn the new/unseen classes to expand the existing model. This is called open-world learning (OWL). This paper proposes a new OWL method based on meta-learning. The key novelty is that the model maintains only a dynamic set of seen classes that allows new classes to be added or deleted with no need for model re-training. Each class is represented by a small set of training examples. In testing, the meta-classifier only uses the examples of the maintained seen classes (including the newly added classes) on-the-fly for classification and rejection. Experimental results with e-commerce product classification show that the proposed method is highly effective1.

Subfields: Artificial Intelligence, Meta & Continual Learning

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

In Meta & Continual Learning

Citations in the corpus

Co-Transport for Class-Incremental Learning (2021, background): “For example, in online opinion monitoring, new topics often emerge as news happens [28], and in the electronic commerce platform, new types of products appear daily [27, 49].”
Learngene: From Open-World to Your Learning Task (2021, background): “Consequently, static and dedicated neural networks cannot resolve incrementally learning the unknown/unseen classes (Fei, Wang, and Liu 2016; Xu et al. 2019) without retraining from scratch and without catastrophic forgetting (French 1999), which is exactly what open-world learning (Bendale and…”
Towards Open World Object Detection (2021, background): “[47] studies open world face identity learning while [64] proposed to use an exemplar set of seen classes to match them against a new sample, and rejects it in case of a low match with all previously known classes.”
Out-of-Domain Detection for Natural Language Understanding in Dialog Systems (2019, neutral): “Note that we did not choose the recently proposed KNN-based model (Xu et al., 2019) as a baseline because the time and memory usage for this model increases dramatically with the size of training data, which is not suitable for large scale real-world NLU systems.”
Learning on the Job: Online Lifelong and Continual Learning (2020, background): “Several researchers have studied this problem, e.g., Scheirer et al. (2013), Bendale and Boult, 2015; Fei and Liu (2016), De Rosa et al (2016), and Xu et al. (2019).”
PyCIL: a Python toolbox for class-incremental learning (2021, background): “For example, when training a robot in the open-world, it meets new objects as time goes, and in the electronic commerce platform, new types of products appear daily (Luo et al., 2020; Xu et al., 2019).”
CG-BERT: Conditional Text Generation with BERT for Generalized Few-shot Intent Detection (2020, background): “Xu et al. (2019) proposes an open-world learning model to deal with the unseen classes in the product classi-ﬁcation problem.”
Composed Variational Natural Language Generation for Few-shot Intents (2020, background): “Most existing works (Hu et al., 2009; Xu and Sarikaya, 2013; Chen et al., 2016; Xia et al., 2018) formulate intent detection as a classification task and achieve high performance on pre-defined intents with sufficient labeled examples.”
Large Language Models on Graphs: A Comprehensive Survey (2023, methodology): “It is important to utilize both the text information and the graph structure information (user behavior) to model users and items and solve complex downstream tasks ( e.g. , item recommendation [106], bundle recommendation [107], and product understanding [108]).”
Generalized Out-of-Distribution Detection: A Survey (2021, unknown): “The difference between DA and DG is that while the former requires extra but few training samples from the target domain, the latter does not.”
Induction Networks for Few-Shot Text Classification (2019, background): “Xu et al. (2019) proposed a open-world learning model to deal with the unseen classes in the product classification problem.”
CLEAR: Contrastive-Prototype Learning with Drift Estimation for Resource Constrained Stream Mining (2021, methodology): “In this paper, we compare the CLEAR with three major strategies: exemplar storage based (iCaRL, L2AC and FSCIL), and parameter regularization based (EWC, LwF and MAS).”
Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference (2020, methodology): “…Few-shot classiﬁcation (Fei-Fei et al., 2006; Vinyals et al., 2016b) has been applied to text classiﬁcation tasks (Deng et al., 2019; Geng et al., 2019; Xu et al., 2019), and few-shot intent detection is also studied but without OOS (Luo et al., 2018; Xia et al., 2020; Casanueva et al., 2020).”
Self-Initiated Open World Learning for Autonomous AI Agents (2021, background): “Yet, some others have studied learning the novel objects or classes after they have been detected (Bendale and Boult 2015; Fei, Wang, and Liu 2016; Xu et al. 2019).”
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; Perera et al. 2020; Miller et al. 2021) and try to solve the problem from different perspectives.”
AI Autonomy: Self-initiated Open-world Continual Learning and Adaptation (2022, background): “Openworld learning has been studied bymany researchers (Bendale and Boult 2015; Fei, Wang, and Liu 2016; Xu et al. 2019), but they mainly focused on novelty detection (Parmar et al.”
HRN: A Holistic Approach to One Class Learning (2020, methodology): “Various forms of thresholding were used to detect anomalies [23, 24, 33, 51, 18, 76, 36, 20, 85].”