Incremental Adaptive Learning Vector Quantization for Character Recognition with Continuous Style Adaptation
Incremental learning enables continuous model adaptation based on a constantly arriving data stream. It is a way relevant to human cognitive system, which learns to predict objects in a changing world. Incremental learning for character recognition is a typical scenario that characters appear sequentially and the font/writing style changes irregularly. In the paper, we investigate how to classify characters incrementally (i.e., input patterns appear once at a time). A reasonable assumption is that adjacent characters from the same font or the same writer share the same style in a short period while style variation occurs in characters printed by different fonts or written by different persons during a long period. The challenging issue here is how to take advantage of the local style consistency and adapt to the continuous style variation as well incrementally. For this purpose, we propose a continuous incremental adaptive learning vector quantization (CIALVQ) method, which incrementally learns a self-adaptive style transfer matrix for mapping input patterns from style-conscious space onto style-free space. After style transformation, this problem is casted into a common character recognition task and an incremental learning vector quantization (ILVQ) classifier is used. In this framework, we consider two learning modes: supervised incremental learning and active incremental learning. In the latter mode, samples receiving low confidence from the classifier are requested class labels. We evaluated the classification performance of CIALVQ in two scenarios, interleaved test-then-train and style-specific classification on NIST hand-printed data sets. The results show that local style consistency improves the accuracies of both two test scenarios, and for both supervised and active incremental learning modes.
KeywordsContinuous incremental adaptive learning vector quantization Style transfer mapping Local style consistency Active learning
Compliance with Ethical Standards
This work has been supported in part by the Strategic Priority Research Program of the CAS Grant XDB02060009 and the National Natural Science Foundation of China (NSFC) Grant 61411136002.
Conflict of Interests
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants or animals performed by any of the authors.
- 1.Bransford JD, Brown AL, Cocking RR. How people learn: brain, mind, experience, and school. National Academy Press. 2000.Google Scholar
- 5.Gong C, Tao D, Yang J, Liu W. Teaching-to-learn and learning-to-teach for multi-label propagation[C]. AAAI. 2016. p. 1610–16.Google Scholar
- 6.Gong C. Exploring commonality and individuality for multi-modal curriculum learning[C]. AAAI. 2017. p. 1926–33.Google Scholar
- 7.Syed N, Liu H, Sung K. Incremental learning with support vector machines[C]. In: International joint conference on artificial intelligence. Sweden: Morgan Kaufmann Publishers. 1999. p. 352–6.Google Scholar
- 8.Hoi SCH, Wang J, Zhao P. Libol: A library for online learning algorithms[J]. J Mach Learn Res. 2014; 15(1):495–9.Google Scholar
- 15.Kohonen T. Improved versions of learning vector quantization[C]. In: International joint conference on neural networks. 1990. p. 545–550.Google Scholar
- 18.Shen YY, Liu CL. Incremental learning vector quantization for character recognition with local style consistency[C]. In: Proceeding of the 8th international conference in brain inspired cognitive systems. 2016. p. 228–39.Google Scholar
- 20.Oza NC. Online bagging and boosting[C]. IEEE International Conference on Systems, Man and Cybernetics:2340–45. 2005.Google Scholar
- 21.Liu X, Yu T. Gradient feature selection for online boosting[C]. ICCV. 2007. p. 18.Google Scholar
- 22.Saffari A, Leistner C, Santner J, et al. On-line random forests[C] In: Computer vision workshops (ICCV Workshops). 2009. p. 1393–400.Google Scholar
- 26.Crammer K, Dredze M, Kulesza A. Multi-class confidence weighted algorithms[C]. In: Proceedings of the conference on empirical methods in natural language processing. 2009. p. 496–504.Google Scholar
- 27.Ushiku Y, Hidaka M, Harada T. Three guidelines of online learning for large-scale visual recognition[C]. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2014. p. 3574–81.Google Scholar
- 28.Gama J, žliobaitė I, Bifet A, Pechenizkiy M, Bouchachia A. A survey on concept drift adaptation[J]. ACM Comput Surv 2014;46(4).Google Scholar
- 31.Huang Z, Ding K, Jin L, et al. Writer adaptive online handwriting recognition using incremental linear discriminant analysis[C]. In: IEEE Proceedings of the conference on document analysis and recognition. 2009. p. 91–5.Google Scholar
- 32.Ding K, Jin L. Incremental MQDF learning for writer adaptive handwriting recognition[C]. In: Proceedings of the conference on frontiers in handwriting recognition (ICFHR). 2010. p. 559– 64.Google Scholar
- 33.Bishop CM. Pattern recognition and machine learning. New York: Springer; 2006.Google Scholar
- 35.Grother PJ. Handprinted forms and character database, NIST special database 19. Technical Report and CDROM. 1995.Google Scholar
- 37.Duchi J, Hazan E, Singer Y. Adaptive subgradient methods for online learning and stochastic optimization[J]. J Mach Learn Res. 2011;12:2121–59.Google Scholar