参考文献:1. Aoki, T., Aoyagi, T.: Self-organizing maps with asymmetric neighborhood function. Neural Computation?19, 2515-535 (2007) CrossRef 2. Aoki, T., Ota, K., Kurata, K., Aoyagi, T.: Ordering process of self-organizing maps improved by asymmetric neighborhood function. Cognitive Neurodyamics?3, 9-5 (2009) CrossRef 3. Ota, K., Aoki, T., Kurata, K., Aoyagi, T.: Asymmetric neighborhood functions accelerate ordering process of self-organizing maps. Physical Review E?83, 021903 (2011) 4. Bauer, H.-U., Villmann, T.: Growing a hypercubical output space in a self-organizing feature map. IEEE Transaction on Neural Networks?8(2), 218-26 (1997) CrossRef 5. Baum, L.E., Petrie, T.: Statistical inference for finite state Markov chains. The Annals of Mathematical Statistics?37(6), 1554-563 (1966) CrossRef 6. Berglund, E., Sitte, J.: The parameter-less self-organizing map algorithm. IEEE Transaction on Neural Networks?17(2), 305-16 (2006) CrossRef 7. Kohonen, T.: Self-Organizing Maps. Series in Information Sciences. Springer, Heidelberg (1995) CrossRef 8. Kuremoto, T., Hano, T., Kobayashi, K., Obayashi, M.: For partner robots: A hand instruction learning system using transient-SOM. In: Proceedings of the 2nd International Conference on Natural Computation and the 3rd International Conference on Fuzzy Systems and Knowledge Discovery (ICNC 2006-FSKD 2006), pp. 403-14 (2006) 9. Kuremoto, T., Obayashi, M., Kobayashi, K., Feng, L.-B.: Instruction learning systems for partner robots. In: Advances in Robotics-Modeling, Control, and Applications, iConcept, ch. 8 (2012) 10. Kuremoto, T., Otani, T., Feng, L.-B., Kobayashi, K., Obayashi, M.: A hand image instruction learning system using PL-G-SOM. In: Proceedings of the 12th International Conference on Artificial Intelligence (ICAI 2012), CD-ROM (2012) 11. Sherrah, S., Gong, S.: Skin Colour Analysis (2001), http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/GONG1/cvOnline-skinColourAnalysis.html 12. Sutton, S.S., Barto, A.G.: Reinforcement Learning: An Instruction. The MIT Press, London (1998)
18. Graduate School of Science and Engineering, Yamaguchi University, Tokiwadai 2-16-1, Ube, Yamaguchi, 755-8611, Japan 19. School of Information Science and Technology, Aichi Prefectural University, Ibaragabasama 1522-3, Nagakute-Shi, Aichi, 480-1198, Japan
ISSN:1611-3349
文摘
In this paper, we adopt an asymmetric neighborhood function proposed by Aoki and Aoyagi in to a PL-G-SOM to improve the learning performance of the hand shape instruction perspective and learning system. The asymmetric neighborhood function was used in a normal SOM and few applications can be found. The novel PL-G-SOM and its improved version are named as "AGSOM" and “IAGSOM-respectively. The effectiveness of the proposed method was confirmed by the experiments with 8 kinds of instructions.