S2OSC: A Holistic Semi-Supervised Approach for Open Set Classification

by   Yang Yang, et al.

Open set classification (OSC) tackles the problem of determining whether the data are in-class or out-of-class during inference, when only provided with a set of in-class examples at training time. Traditional OSC methods usually train discriminative or generative models with in-class data, then utilize the pre-trained models to classify test data directly. However, these methods always suffer from embedding confusion problem, i.e., partial out-of-class instances are mixed with in-class ones of similar semantics, making it difficult to classify. To solve this problem, we unify semi-supervised learning to develop a novel OSC algorithm, S2OSC, that incorporates out-of-class instances filtering and model re-training in a transductive manner. In detail, given a pool of newly coming test data, S2OSC firstly filters distinct out-of-class instances using the pre-trained model, and annotates super-class for them. Then, S2OSC trains a holistic classification model by combing in-class and out-of-class labeled data and remaining unlabeled test data in semi-supervised paradigm, which also integrates pre-trained model for knowledge distillation to further separate mixed instances. Despite its simplicity, the experimental results show that S2OSC achieves state-of-the-art performance across a variety of OSC tasks, including 85.4 pseudo-labels. We also demonstrate how S2OSC can be expanded to incremental OSC setting effectively with streaming data.



There are no comments yet.


page 16

page 18


When Semi-Supervised Learning Meets Transfer Learning: Training Strategies, Models and Datasets

Semi-Supervised Learning (SSL) has been proved to be an effective way to...

Open-World Semi-Supervised Learning

Supervised and semi-supervised learning methods have been traditionally ...

Towards Semi-Supervised Learning for Deep Semantic Role Labeling

Neural models have shown several state-of-the-art performances on Semant...

Detecting Extrapolation with Local Ensembles

We present local ensembles, a method for detecting extrapolation at test...

Dual-Teacher Class-Incremental Learning With Data-Free Generative Replay

This paper proposes two novel knowledge transfer techniques for class-in...

Contrastive Model Inversion for Data-Free Knowledge Distillation

Model inversion, whose goal is to recover training data from a pre-train...

A Novel DNN Training Framework via Data Sampling and Multi-Task Optimization

Conventional DNN training paradigms typically rely on one training set a...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.