Access the full text.
Sign up today, get DeepDyve free for 14 days.
Zhuoyi Wang, Zelun Kong, Swarup Chandra, Hemeng Tao, L. Khan (2019)
Robust High Dimensional Stream Classification with Novel Class Detection2019 IEEE 35th International Conference on Data Engineering (ICDE)
Geoffrey Hinton, O. Vinyals, J. Dean (2015)
Distilling the Knowledge in a Neural NetworkArXiv, abs/1503.02531
Sylvestre-Alvise Rebuffi, Alexander Kolesnikov, G. Sperl, Christoph Lampert (2016)
iCaRL: Incremental Classifier and Representation Learning2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Roger Ratcliff (1990)
Connectionist models of recognition memory: constraints imposed by learning and forgetting functionsConnectionist models of recognition memory: constraints imposed by learning and forgetting functions.Psychological Review, 97
B. Scholkopf, John Platt, J. Shawe-Taylor, Alex Smola, R. Williamson (2001)
Estimating the Support of a High-Dimensional DistributionNeural Computation, 13
Arslan Chaudhry, P. Dokania, Thalaiyasingam Ajanthan, Philip Torr (2018)
Riemannian Walk for Incremental Learning: Understanding Forgetting and IntransigenceArXiv, abs/1801.10112
J. Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, J. Veness, Guillaume Desjardins, Andrei Rusu, Kieran Milan, John Quan, Tiago Ramalho, A. Grabska-Barwinska, D. Hassabis, C. Clopath, D. Kumaran, R. Hadsell (2016)
Overcoming catastrophic forgetting in neural networksProceedings of the National Academy of Sciences, 114
Yan Xia, Xudong Cao, Fang Wen, G. Hua, Jian Sun (2015)
Learning Discriminative Reconstructions for Unsupervised Outlier Removal2015 IEEE International Conference on Computer Vision (ICCV)
ZongYuan Ge, S. Demyanov, Zetao Chen, R. Garnavi (2017)
Generative OpenMax for Multi-Class Open Set ClassificationArXiv, abs/1707.07418
R. Ratcliff (1990)
Connectionist models of recognition memory: constraints imposed by learning and forgetting functions.Psychological review, 97 2
E. Cubuk, Barret Zoph, Dandelion Mané, Vijay Vasudevan, Quoc Le (2019)
AutoAugment: Learning Augmentation Strategies From Data2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Terrance Devries, Graham Taylor (2017)
Improved Regularization of Convolutional Neural Networks with CutoutArXiv, abs/1708.04552
Yen-Chang Hsu, Zhaoyang Lv, Joel Schlosser, Phillip Odom, Z. Kira (2019)
Multi-class Classification without Multi-class LabelsArXiv, abs/1901.00544
Yann LeCun, Corinna Cortes, Christopher J. C. Burges (1998)
The MNIST database of handwritten digits, 1998Retrieved from http://yann.lecun.com/exdb/mnist. 10 (1998), 34., 10
Dan Hendrycks, Kevin Gimpel (2016)
A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural NetworksArXiv, abs/1610.02136
Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi (2018)
Open Set Recognition by Regularising Classifier with Fake Data Generated by Generative Adversarial Networks2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Xin Mu, Feida Zhu, Juan Du, Ee-Peng Lim, Zhi-Hua Zhou (2017)
Streaming Classification with Emerging New Class by Class Matrix Sketching
V. Vovk, Ruodu Wang (2012)
Combining P-Values Via AveragingPolitical Methods: Quantitative Methods eJournal
Kihyuk Sohn, David Berthelot, Chun-Liang Li, Zizhao Zhang, Nicholas Carlini, E. Cubuk, Alexey Kurakin, Han Zhang, Colin Raffel (2020)
FixMatch: Simplifying Semi-Supervised Learning with Consistency and ConfidenceArXiv, abs/2001.07685
Ilya Sutskever, James Martens, George Dahl, Geoffrey Hinton (2013)
On the importance of initialization and momentum in deep learning
Soravit Changpinyo, Wei-Lun Chao, Boqing Gong, Fei Sha (2016)
Synthesized Classifiers for Zero-Shot Learning2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Shiyu Liang, Yixuan Li, R. Srikant (2017)
Enhancing The Reliability of Out-of-distribution Image Detection in Neural NetworksarXiv: Learning
Fei Liu, K. Ting, Zhi-Hua Zhou (2008)
Isolation Forest2008 Eighth IEEE International Conference on Data Mining
Yuval Netzer, Tao Wang, Adam Coates, A. Bissacco, Bo Wu, A. Ng (2011)
Reading Digits in Natural Images with Unsupervised Feature Learning
Ge Song, Xiaoyang Tan (2019)
Sequential Learning for Cross-Modal Retrieval2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)
Jingjing Li, Mengmeng Jing, Ke Lu, Lei Zhu, Yang Yang, Zi Huang (2019)
Alleviating Feature Confusion for Generative Zero-shot LearningProceedings of the 27th ACM International Conference on Multimedia
Lawrence Neal, Matthew Olson, Xiaoli Fern, Weng-Keen Wong, Fuxin Li (2018)
Open Set Learning with Counterfactual Images
Qing Da, Yang Yu, Zhi-Hua Zhou (2014)
Learning with Augmented Class by Exploiting Unlabeled Data
Xin-Qiang Cai, Peng Zhao, K. Ting, Xin Mu, Yuan Jiang (2019)
Nearest Neighbor Ensembles: An Effective Method for Difficult Problems in Streaming Classification with Emerging New Classes2019 IEEE International Conference on Data Mining (ICDM)
David Berthelot, Nicholas Carlini, E. Cubuk, Alexey Kurakin, Kihyuk Sohn, Han Zhang, Colin Raffel (2020)
ReMixMatch: Semi-Supervised Learning with Distribution Matching and Augmentation Anchoring
Han Xiao, Kashif Rasul, Roland Vollgraf (2017)
Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning AlgorithmsArXiv, abs/1708.07747
Kaiming He, X. Zhang, Shaoqing Ren, Jian Sun (2015)
Deep Residual Learning for Image Recognition2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
B. Kulis (2013)
Metric Learning: A SurveyFound. Trends Mach. Learn., 5
Sang-Woo Lee, Jin-Hwa Kim, Jaehyun Jun, Jung-Woo Ha, Byoung-Tak Zhang (2017)
Overcoming Catastrophic Forgetting by Incremental Moment MatchingArXiv, abs/1703.08475
Chuanxing Geng, Sheng-Jun Huang, Songcan Chen (2018)
Recent Advances in Open Set Recognition: A SurveyIEEE Transactions on Pattern Analysis and Machine Intelligence, 43
Ki Kim, Sang-Heon Shim, Yongsub Lim, Jongseob Jeon, Jeong-Woo Choi, Byungchan Kim, A. Yoon (2020)
RaPP: Novelty Detection with Reconstruction along Projection Pathway
L. Maaten, Geoffrey Hinton (2008)
Visualizing Data using t-SNEJournal of Machine Learning Research, 9
K. Han, A. Vedaldi, Andrew Zisserman (2019)
Learning to Discover Novel Visual Categories via Deep Transfer Clustering2019 IEEE/CVF International Conference on Computer Vision (ICCV)
Weiyang Liu, Yandong Wen, Zhiding Yu, Meng Yang (2016)
Large-Margin Softmax Loss for Convolutional Neural Networks
Yann LeCun, Corinna Cortes (2005)
The mnist database of handwritten digits
A. Krizhevsky (2009)
Learning Multiple Layers of Features from Tiny Images
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 the owned in-class data, and then utilize the pre-trained models to classify test data directly. However, these methods always suffer from the 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, which 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 the mostly 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 with the remaining unlabeled test data in a semi-supervised paradigm. Furthermore, considering that data are usually in the streaming form in real applications, we extend S2OSC into an incremental update framework (I-S2OSC), and adopt a knowledge memory regularization to mitigate the catastrophic forgetting problem in incremental update. Despite the simplicity of proposed models, the experimental results show that S2OSC achieves state-of-the-art performance across a variety of OSC tasks, including 85.4% of F1 on CIFAR-10 with only 300 pseudo-labels. We also demonstrate how S2OSC can be expanded to incremental OSC setting effectively with streaming data.
ACM Transactions on Knowledge Discovery from Data (TKDD) – Association for Computing Machinery
Published: Sep 4, 2021
Keywords: Open set classification
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.