Online active classification via margin-based and feature-based label queries
Abstract
In the paradigm of online active classification, the learner not only has to predict the label of each incoming instance, but also must decide whether the true label of that instance should be supplied, or not. The overall goal is to minimize the number of prediction mistakes with few label queries. In this paper, we focus on a novel framework for online active learning, with the aim of handling high dimensional classification problems. The key component of our framework is to exploit both the margin-based predictive uncertainty and the feature-based discriminative information of the current instance, in order to determine whether it should be labeled. Based on this labeling strategy, we propose several online active learning algorithms, for both binary classification tasks and multiclass ones. For these algorithms, which use adaptive subgradient methods for updating their linear model, expected mistake bounds are provided. Experiments on high-dimensional (binary and multiclass) classification datasets reveal the benefit of our label query strategy, and show the superiority of our algorithms over the existing ones.
References
[1]
Awasthi, P., Balcan, M., Haghtalab, N., & Urner, R. (2015). Efficient learning of linear separators under bounded noise. In Proceedings of the 28th Conference on Learning Theory, Paris, France, vol 40 (pp. 167–190).
[2]
Balcan, M., & Long, P. M. (2013). Active and passive learning of linear separators under log-concave distributions. In Proceedings of the 26th Annual Conference on Learning Theory, Princeton University, NJ, USA, vol 30 (pp. 288–316).
[3]
Cesa-Bianchi N, Gentile C, and Zaniboni L Worst-case analysis of selective sampling for linear classification Journal of Machine Learning Research 2006 7 1205-1230
[4]
Crammer K, Dekel O, Keshet J, Shalev-Shwartz S, and Singer Y Online passive-aggressive algorithms J Mach Learn Res 2006 7 551-585
[5]
Crammer K, Dredze M, and Pereira F Confidence-weighted linear classification for text categorization Journal of Machine Learning Research 2012 13 1891-1926
[6]
Crammer K, Kulesza A, and Dredze M Adaptive regularization of weight vectors Machine Learning 2013 91 2 155-187
[7]
Demir B and Bruzzone L A multiple criteria active learning method for support vector regression Pattern Recognition 2014 47 7 2558-2567
[8]
Du B, Wang Z, Zhang L, Zhang L, Liu W, Shen J, and Tao D Exploring representativeness and informativeness for active learning IEEE Transactions on Cybernetics 2017 47 1 14-26
[9]
Duchi JC, Hazan E, and Singer Y Adaptive subgradient methods for online learning and stochastic optimization Journal of Machine Learning Research 2011 12 2121-2159
[10]
Golovin, D., Krause, A., & Ray, D. (2010). Near-optimal bayesian active learning with noisy observations. In Advances in Neural Information Processing Systems (pp. 766–774).
[11]
Hanneke S Theory of disagreement-based active learning Foundations and Trends in Machine Learning 2014 7 2–3 131-309
[12]
Hao S, Lu J, Zhao P, Zhang C, Hoi SCH, and Miao C Second-order online active learning and its applications IEEE Transactions on Knowledge and Data Engineering 2018 30 7 1338-1351
[13]
Hazan E, Agarwal A, and Kale S Logarithmic regret algorithms for online convex optimization Machine Learning 2007 69 2–3 169-192
[14]
Hoi SCH, Jin R, Zhao P, and Yang T Online multiple kernel classification Machine Learning 2013 90 2 289-316
[15]
Huang S, Jin R, and Zhou Z Active learning by querying informative and representative examples IEEE Transactions on Pattern Analysis and Machine Intelligence 2014 36 10 1936-1949
[16]
Katakis I, Tsoumakas G, Banos E, Bassiliades N, and Vlahavas IP An adaptive personalized news dissemination system Journal of Intelligent Information Systems 2009 32 2 191-212
[17]
Lu J, Hoi SCH, Wang J, Zhao P, and Liu Z Large scale online kernel learning Journal of Machine Learning Research 2016 17 47:1-47:43
[18]
Lu J, Zhao P, and Hoi SCH Online passive-aggressive active learning Machine Learning 2016 103 2 141-183
[19]
Lughofer E On-line active learning: A new paradigm to improve practical useability of data stream modeling methods Information Sciences 2017 415 356-376
[20]
Luo, H., Agarwal, A., Cesa-Bianchi, N., & Langford, J. (2016). Efficient second order online learning by sketching. In Advances in Neural Information Processing Systems (pp. 902–910).
[21]
Ma, J., Saul, L. K., Savage, S., & Voelker, G. M. (2009). Identifying suspicious urls: an application of large-scale online learning. In Proceedings of the 26th International Conference on Machine Learning, Montreal, Quebec, Canada (pp. 681–688).
[22]
Settles, B. (2009). Active learning literature survey. Computer Sciences Technical Report 1648, University of Wisconsin-Madison.
[23]
Shalev-Shwartz S Online learning and online convex optimization Foundations and Trends in Machine Learning 2012 4 2 107-194
[24]
Shalev-Shwartz S, Singer Y, Srebro N, and Cotter A Pegasos: Primal estimated sub-gradient solver for SVM Mathematical Programming 2011 127 1 3-30
[25]
Song Q, Xu Z, Fan H, and Wang D Robust recurrent kernel online learning IEEE Transactions on Neural Networks and Learning Systems 2017 28 5 1068-1081
[26]
Sun Y, Tang K, Minku LL, Wang S, and Yao X Online ensemble learning of data streams with gradually evolved classes IEEE Transactions on Knowledge and Data Engineering 2016 28 6 1532-1545
[27]
Tosh, C., & Dasgupta, S. (2017). Diameter-based active learning. In Proceedings of the 34th International Conference on Machine Learning, vol 70 (pp. 3444–3452).
[28]
Wang Z and Ye J Querying discriminative and representative samples for batch mode active learning ACM Transactions on Knowledge Discovery from Data 2015 9 3 17:1-17:23
[29]
Zhai T, Gao Y, Wang H, and Cao L Classification of high-dimensional evolving data streams via a resource-efficient online ensemble Data Mining and Knowledge Discovery 2017 31 5 1242-1265
[30]
Zhai T, Koriche F, Wang H, and Gao Y Tracking sparse linear classifiers IEEE Transactions on Neural Networks and Learning Systems 2019 30 7 2079-2092
[31]
Zhang, C. (2018). Efficient active learning of sparse halfspaces. In Proceeding of the 31st Conference on Learning Theory, Stockholm, Sweden, vol 75 (pp. 1856–1880).
[32]
Zhao, P., & Hoi, S. C. H. (2013). Cost-sensitive online active learning with application to malicious URL detection. In Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Chicago, IL, USA (pp. 919–927).
[33]
Zinkevich, M. (2003). Online convex programming and generalized infinitesimal gradient ascent. In Proceedings of the 20th International Conference on Machine Learning, Washington, DC, USA (pp. 928–936).
Recommendations
An efficient online active learning algorithm for binary classification
We propose a new online active learning algorithm for binary classification.Our algorithm uses a margin-based criterion with iteratively decreased threshold.Our algorithm requires less queries to achieve comparable classification accuracy.Our algorithm ...
Online Multi-label Passive Aggressive Active Learning Algorithm Based on Binary Relevance
Neural Information ProcessingAbstractOnline multi-label learning is an efficient classification paradigm in machine learning. However, traditional online multi-label methods often need requesting all class labels of each incoming sample, which is often human cost and time-consuming ...
Missing multi-label learning with non-equilibrium based on classification margin
AbstractMulti-labels are more suitable for the ambiguity of the real world. However, missing labels are common in multi-label learning datasets; this results in unbalanced labeling and label diversity, which directly affect the performance of ...
Highlights- The classification margin is proposed to expand the label space by the label density, which aims to reduce the influence of threshold function on labeling ...
Comments
Information & Contributors
Information
Published In
© The Author(s), under exclusive licence to Springer Science+Business Media LLC, part of Springer Nature 2022.
Publisher
Kluwer Academic Publishers
United States
Publication History
Published: 01 June 2022
Accepted: 06 February 2022
Revision received: 22 December 2021
Received: 26 January 2021
Author Tags
Qualifiers
- Research-article
Funding Sources
- National Natural Science Foundation of China
- Natural Science Foundation of the Jiangsu Higher Education Institutions of China
- national natural science foundation of china
- National Natural Science Foundation of China
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 28 Dec 2024