Learning Vector Quantization (LVQ) has been intensively studied to generate good reference models in pattern recognition since 1986, and it has some nice theoretical properties. However, the design of reference models based on LVQ suffers from several major drawbacks for the recognition of large-set patterns, in which good reference models play an important role in achieving high performance. They are due in large part to the following facts: (1) it may not generate good reference models, if the initial values of the reference models are outside the convex hull of the input data, (2) it cannot guarantee optimal reference models due to the strategy to accept new reference models in each iteration step, and (3) it is apt to get stuck at overtraining phenomenon. In this paper, we first discuss the impact of these problems. And then, to cope with these, we propose a new method for the optimal design of large-set reference models using an improved LVQ3 combined with Simulated Annealing which has been proven to be a useful technique in many areas of optimization problems. Experimental results with large-set handwritten characters reveal that the proposed method is superior to the conventional method based on averaging and other LVQ-based methods.
Bibliographical noteFunding Information:
Acknowledgements: This research was supported by the 1992 Directed Basic Research Fund of Korea Science and Engineering Foundation. Requests for reprints should be sent to S.-W. Lee, Department of Computer Science, Korea University, Seongbuk-ku, Seoul 136--701 Korea, the standard K-means algorithm, LBG algorithm (Linde, Buzo, & Gray, 1980), Fuzzy VQ algorithm (Selim & Ismail, 1984), have been well developed and widely applied in various fields. These methods try to minimize the average distortion error through an optimal choice of reference models.
- large-set pattern recognition
- learning vector quantization
- optimal design of large-set reference models
- simulated annealing
- vector quantization
ASJC Scopus subject areas
- Cognitive Neuroscience
- Artificial Intelligence