{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,12]],"date-time":"2025-10-12T04:57:41Z","timestamp":1760245061196},"reference-count":57,"publisher":"MIT Press","issue":"7","content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2021,6,11]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>This letter focuses on the problem of lifelong classification in the open world, the goal of which is to achieve an endless process of learning. However, incremental data sets (like the streaming data) in the open world, where the new classes may be emerging, are unsuited for classical classification methods. For addressing this problem, existing methods usually retrain the whole observed data sets with the complex computation and the expensive storage cost. This letter attempts to improve the performance of classification in the open world and decomposes the problem into three subproblems: (1) to reject unknown instances, (2) to classify accepted instances, and (3) to cut the cost of learning. Rejecting unknown instances refers to recognize those instances whose classes are unknown according to the learner, which could reduce the computation of the retraining process and eliminate the storage of historical data sets. We employ outlier detection for rejecting instances and a variant artificial neural network for classifying with fewer weights. Results on several experiments show that the work is effective. Source code can be found at https:\/\/github.com\/wangbi1988\/Lifelong-learning-in-Open-World-Classification.<\/jats:p>","DOI":"10.1162\/neco_a_01391","type":"journal-article","created":{"date-parts":[[2021,4,22]],"date-time":"2021-04-22T23:21:38Z","timestamp":1619133698000},"page":"1818-1852","update-policy":"http:\/\/dx.doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":1,"title":["Lifelong Classification in Open World With Limited Storage Requirements"],"prefix":"10.1162","volume":"33","author":[{"given":"Wang","family":"Bi","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Southeast University, Nanjing 210000, China wangbi@seu.edu.cn"}]},{"given":"Chen","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China shawn.cy@nuaa.edu.cn"}]},{"given":"Li","family":"XueLian","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Southeast University, Nanjing 210000, China 230159108@seu.edu.cn"}]},{"given":"Chen","family":"JunFu","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China cjf@nuaa.edu.cn"}]}],"member":"281","published-online":{"date-parts":[[2021,6,11]]},"reference":[{"key":"2021101521491338400_B1","first-page":"7120","article-title":"Expert gate: Lifelong learning with a network of experts","author":"Aljundi","year":"2017","journal-title":"Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"2021101521491338400_B2","first-page":"643","article-title":"A general evaluation measure for document organization tasks","author":"Amig\u00f3","year":"2013","journal-title":"Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval"},{"key":"2021101521491338400_B3","author":"Bartlett","year":"2017","journal-title":"Nearly-tight VC-dimension and pseudodimension bounds for piecewise linear neural networks."},{"key":"2021101521491338400_B4","first-page":"325:992","article-title":"Learning to continually learn","author":"Beaulieu","year":"2020","journal-title":"Proceedings of the 24th European Conference on Artificial Intelligence"},{"key":"2021101521491338400_B5","first-page":"1893","article-title":"Towards open world recognition","author":"Bendale","year":"2015","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"2021101521491338400_B6","first-page":"82","volume-title":"Advances in neural information processing systems","author":"Bengio","year":"2009"},{"key":"2021101521491338400_B7","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1145\/1553374.1553380","article-title":"Curriculum learning","author":"Bengio","year":"2009","journal-title":"Proceedings of the 26th Annual International Conference on Machine Learning"},{"key":"2021101521491338400_B8","doi-asserted-by":"crossref","first-page":"93","DOI":"10.1145\/335191.335388","article-title":"LOF: Identifying density-based local outliers","volume":"29","author":"Breunig","year":"2000","journal-title":"ACM Sigmod Record"},{"key":"2021101521491338400_B9","first-page":"970","article-title":"Nearest neighbor ensembles: An effective method for difficult problems in streaming classification with emerging new classes","author":"Cai","year":"2019","journal-title":"Proceedings of the IEEE International Conference on Data Mining"},{"key":"2021101521491338400_B10","first-page":"3","article-title":"Toward an architecture for never-ending language learning","author":"Carlson","year":"2010","journal-title":"Proceedings of the 24th AAAI Conference on Artificial Intelligence"},{"issue":"1","key":"2021101521491338400_B11","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1023\/A:1007379606734","article-title":"Multitask learning","volume":"28","author":"Caruana","year":"1997","journal-title":"Machine Learning"},{"issue":"3","key":"2021101521491338400_B12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2200\/S00737ED1V01Y201610AIM033","article-title":"Lifelong machine learning","volume":"10","author":"Chen","year":"2016","journal-title":"Synthesis Lectures on Artificial Intelligence and Machine Learning"},{"key":"2021101521491338400_B13","author":"Chen","year":"2018","journal-title":"Lifelong learning for sentiment classification."},{"key":"2021101521491338400_B14","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1109\/TIT.1970.1054406","article-title":"On optimum recognition error and reject tradeoff","volume":"16","author":"Chow","year":"1970","journal-title":"IEEE Transactions on Information Theory"},{"key":"2021101521491338400_B15","first-page":"1760","article-title":"Learning with augmented class by exploiting unlabeled data","volume":"3","author":"Da","year":"2014","journal-title":"Proceedings of the National Conference on Artificial Intelligence"},{"issue":"7612","key":"2021101521491338400_B16","doi-asserted-by":"crossref","first-page":"420","DOI":"10.1038\/nature18630","article-title":"Prefrontal neuronal assemblies temporally control fear behaviour","volume":"535","author":"Dejean","year":"2016","journal-title":"Nature"},{"key":"2021101521491338400_B17","doi-asserted-by":"crossref","first-page":"1565","DOI":"10.1145\/2939672.2939835","article-title":"Learning cumulatively to become more knowledgeable","author":"Fei","year":"2016","journal-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining"},{"key":"2021101521491338400_B18","article-title":"Knowledge transfer in learning to recognize visual objects classes.","author":"Fei-Fei","year":"2006","journal-title":"Proceedings of the Fifth International Conference on Development and Learning."},{"issue":"8","key":"2021101521491338400_B19","article-title":"Recent advances in open set recognition: A survey","volume":"14","author":"Geng","year":"2020","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"5786","key":"2021101521491338400_B20","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"2021101521491338400_B21","author":"Hinton","year":"2012","journal-title":"Improving neural networks by preventing co-adaptation of feature detectors"},{"key":"2021101521491338400_B22","author":"Isele","year":"2018","journal-title":"Selective experience replay for lifelong learning."},{"key":"2021101521491338400_B23","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1162\/neco.1991.3.1.79","article-title":"Adaptive mixture of experts","volume":"3","author":"Jacobs","year":"1991","journal-title":"Neural Computation"},{"key":"2021101521491338400_B24","doi-asserted-by":"crossref","first-page":"804","DOI":"10.1016\/j.procs.2017.11.440","article-title":"Concept drift in streaming data classification: Algorithms, platforms and issues","volume":"122","author":"Janardan","year":"2017","journal-title":"Procedia Computer Science"},{"key":"2021101521491338400_B25","first-page":"1","volume-title":"Advances in neural information processing systems","author":"Javed","year":"2019"},{"key":"2021101521491338400_B26","article-title":"A method for stochastic optimization","author":"Kinga","year":"2015","journal-title":"Proceedings of the International Conference on Learning Representations"},{"key":"2021101521491338400_B27","author":"Krizhevsky","year":"2009","journal-title":"Learning multiple layers of features from tiny images"},{"key":"2021101521491338400_B28","author":"Kumar","year":"2012","journal-title":"Learning task grouping and overlap in multi-task learning"},{"issue":"7299","key":"2021101521491338400_B29","doi-asserted-by":"crossref","DOI":"10.1038\/nature09108","article-title":"Global and local FMRI signals driven by neurons defined optogenetically by type and wiring","volume":"465","author":"Lee","year":"2010","journal-title":"Nature"},{"key":"2021101521491338400_B30","first-page":"413","article-title":"Isolation forest.","author":"Liu","year":"2008","journal-title":"Proceedings of the Eighth IEEE International Conference on Data Mining"},{"key":"2021101521491338400_B31","article-title":"Large-scale long-tailed recognition in an open world","author":"Liu","year":"2019","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"2021101521491338400_B32","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"Maaten","year":"2008","journal-title":"Journal of Machine Learning Research"},{"issue":"2","key":"2021101521491338400_B33","doi-asserted-by":"crossref","first-page":"275","DOI":"10.1007\/s10462-012-9338-y","article-title":"Mixture of experts: A literature survey","volume":"42","author":"Masoudnia","year":"2014","journal-title":"Artificial Intelligence Review"},{"issue":"5","key":"2021101521491338400_B34","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1145\/3191513","article-title":"Never-ending learning","volume":"61","author":"Mitchell","year":"2018","journal-title":"Communications of the ACM"},{"issue":"8","key":"2021101521491338400_B35","doi-asserted-by":"crossref","first-page":"1605","DOI":"10.1109\/TKDE.2017.2691702","article-title":"Classification under streaming emerging new classes: A solution using completely-random trees","volume":"29","author":"Mu","year":"2017","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"2021101521491338400_B36","first-page":"11313","article-title":"Learning to remember: A synaptic plasticity driven framework for continual learning","author":"Ostapenko","year":"2019","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"key":"2021101521491338400_B37","first-page":"1540","volume-title":"Advances in neural information processing system","author":"Pentina","year":"2015"},{"key":"2021101521491338400_B38","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1145\/335191.335437","article-title":"Efficient algorithms for mining outliers from large data sets","volume":"29","author":"Ramaswamy","year":"2000","journal-title":"IACM Sigmod Record"},{"issue":"3","key":"2021101521491338400_B39","doi-asserted-by":"crossref","first-page":"212","DOI":"10.1080\/00401706.1999.10485670","article-title":"A fast algorithm for the minimum covariance determinant estimator","volume":"41","author":"Rousseeuw","year":"1999","journal-title":"Technometrics"},{"issue":"3","key":"2021101521491338400_B40","doi-asserted-by":"crossref","DOI":"10.1109\/TPAMI.2017.2707495","article-title":"The extreme value machine","volume":"40","author":"Rudd","year":"2017","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"2021101521491338400_B41","first-page":"507","article-title":"Ella: An efficient lifelong learning algorithm","author":"Ruvolo","year":"2013","journal-title":"Proceedings of the International Conference on Machine Learning"},{"issue":"1","key":"2021101521491338400_B42","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1016\/0097-3165(72)90019-2","article-title":"On the density of families of sets","volume":"13","author":"Sauer","year":"1972","journal-title":"Journal of Combinatorial Theory, Series A"},{"issue":"7","key":"2021101521491338400_B43","doi-asserted-by":"crossref","first-page":"1757","DOI":"10.1109\/TPAMI.2012.256","article-title":"Toward open set recognition","volume":"35","author":"Scheirer","year":"2013","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"7","key":"2021101521491338400_B44","doi-asserted-by":"crossref","first-page":"1443","DOI":"10.1162\/089976601750264965","article-title":"Estimating the support of a high-dimensional distribution","volume":"13","author":"Sch\u00f6lkopf","year":"2001","journal-title":"Neural Computation"},{"key":"2021101521491338400_B45","author":"Shalev-Shwartz","year":"2014","journal-title":"Understanding machine learning: From theory to algorithms"},{"key":"2021101521491338400_B46","author":"Shu","year":"2018","journal-title":"Unseen class discovery in open-world classification"},{"key":"2021101521491338400_B47","article-title":"Lifelong machine learning systems: Beyond learning algorithms","author":"Silver","year":"2013","journal-title":"Proceedings of the AAAI Spring Symposium: Lifelong Machine Learning"},{"key":"2021101521491338400_B48","author":"Simonyan","year":"2014","journal-title":"Very deep convolutional networks for large- scale image recognition."},{"issue":"12","key":"2021101521491338400_B49","doi-asserted-by":"crossref","first-page":"3358","DOI":"10.1016\/j.patcog.2007.04.009","article-title":"Cost-sensitive boosting for classification of imbalanced data","volume":"40","author":"Sun","year":"2007","journal-title":"Pattern Recognition"},{"key":"2021101521491338400_B50","article-title":"A deep hierarchical approach to lifelong learning in minecraft","volume":"3","author":"Tessler","year":"2017","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"2021101521491338400_B51","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1016\/B978-044482250-5\/50015-3","article-title":"A lifelong learning perspective for mobile robot control","author":"Thrun","year":"1995","journal-title":"Proceedings of the IEEE\/RSJ\/GI Conference on Intelligent Robots and Systems"},{"key":"2021101521491338400_B52","first-page":"1058","article-title":"Regularization of neural networks using dropconnect","author":"Wan","year":"2013","journal-title":"Proceedings of the International Conference on Machine Learning"},{"issue":"2","key":"2021101521491338400_B53","doi-asserted-by":"crossref","first-page":"2610","DOI":"10.1109\/LRA.2020.2972879","article-title":"Learning natural locomotion behaviors for humanoid robots using human bias","volume":"5","author":"Yang","year":"2020","journal-title":"IEEE Robotics and Automation Letters"},{"key":"2021101521491338400_B54","first-page":"111","article-title":"Classification methods with reject option based on convex risk minimization","volume":"11","author":"Yuan","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"2021101521491338400_B55","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1016\/j.eswa.2017.06.004","article-title":"A soft computing approach for benign and malicious web robot detection","volume":"87","author":"Zabihimayvan","year":"2017","journal-title":"Expert Systems with Applications"},{"issue":"96","key":"2021101521491338400_B56","first-page":"1","article-title":"Pyod: A Python toolbox for scalable outlier detection","volume":"20","author":"Zhao","year":"2019","journal-title":"Journal of Machine Learning Research"},{"key":"2021101521491338400_B57","author":"Zhong","year":"2017","journal-title":"Random erasing data augmentation"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/direct.mit.edu\/neco\/article-pdf\/33\/7\/1818\/1966852\/neco_a_01391.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/direct.mit.edu\/neco\/article-pdf\/33\/7\/1818\/1966852\/neco_a_01391.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,2]],"date-time":"2023-11-02T15:55:35Z","timestamp":1698940535000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/33\/7\/1818\/100573\/Lifelong-Classification-in-Open-World-With-Limited"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,11]]},"references-count":57,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2021,6,11]]},"published-print":{"date-parts":[[2021,6,11]]}},"URL":"https:\/\/doi.org\/10.1162\/neco_a_01391","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,7]]},"published":{"date-parts":[[2021,6,11]]}}}