This paper proposes a novel neural network architecture based on adaptive resonance theory (ART) called ARTgrid that can perform both online and offline clustering of 2D object structures. The main novelty of the proposed architecture is a two-level categorization and search mechanism that can enhance computation speed while maintaining high performance in cases of higher vigilance values. ARTgrid is developed for specific robotic applications for work in unstructured environments with diverse work objects. For that reason simulations are conducted on random generated data which represents actual manipulation objects, that is, their respective 2D structures. ARTgrid verification is done through comparison in clustering speed with the fuzzy ART algorithm and Adaptive Fuzzy Shadow (AFS) network. Simulation results show that by applying higher vigilance values () clustering performance of ARTgrid is considerably better, while lower vigilance values produce comparable results with the original fuzzy ART algorithm. 1. Introduction Adaptive Resonance Theory (ART) [1] is a cognitive neural theory that attempts to explain how the human brain autonomously learns, categorizes, recognizes, and predicts events in a dynamic and changing environment. ART contains a series of artificial neural networks (ANN), which are used for supervised and unsupervised learning. ART neural networks solve the stability-plasticity dilemma defined by Grossberg [2]. Plasticity of a learning algorithm denotes the characteristic of successful adaptation to changing environmental conditions and the possibility to code new input patterns. Stability of learning algorithms is characterized by the ability to learn new input patterns without catastrophic forgetting [3]. Mechanisms and main principles of Adaptive Resonance Theory can be observed in many areas of the human brain including the visual cortex as noted from significant experiments during the previous decades [4, 5]. Main principles are based on the assumption that learning apart from knowledge update utilizes two major mechanisms: categorization and expectation. The main ART mechanisms that are noted in recent ART based clustering architectures mostly utilize the search, choice, and resonance mechanisms from fuzzy ART. Fuzzy ART [6] enables fast categorization and learning performance of analog input patterns. Long-term connection weight values can only decrease in time which provides fuzzy ART with high clustering stability. The complement coding mechanism ensures stable normalization of input vectors. A fuzzy ART variant algorithm
References
[1]
G. A. Carpenter and S. Grossberg, “Adaptive resonance theory,” in Encyclopedia of Machine Learning, Springer, New York, NY, USA, 2011.
[2]
S. Grossberg, “Adaptive resonance theory: how a brain learns to consciously attend, learn, and recognize a changing world,” Neural Networks, vol. 37, pp. 1–47, 2013.
[3]
M. Hattori, Avoiding Catastrophic Forgetting by a Biologically Inspired Dual-Network Memory Model, vol. 7664, Springer, Berlin, Germany, 2012.
[4]
S. Grossberg, How Does the Cerebral Cortex Work? Developement, Learning, Attention, and 3D Vision by Laminar Circuits of Visual Cortex, Boston University Center for Adaptive Systems, Department of Cognitive and Neural Systems, 2003.
[5]
S. Grossberg and L. R. Pearson, “Laminar cortical dynamics of cognitive and motor working memory, sequence learning and performance: toward a unified theory of how the cerebral cortex works,” Psychological Review, vol. 115, no. 3, pp. 677–732, 2008.
[6]
G. A. Carpenter, S. Grossberg, and D. B. Rosen, “Fuzzy ART: fast stable learning and categorization of analog patterns by an adaptive resonance system,” Neural Networks, vol. 4, no. 6, pp. 759–771, 1991.
[7]
M. Georgiopoulos, I. Dagher, G. L. Heileman, and G. Bebis, “Properties of learning of a fuzzy ART variant,” Neural Networks, vol. 12, no. 6, pp. 837–850, 1999.
[8]
A.-H. Tan, G. A. Carpenter, and S. Grossberg, “Intelligence through interaction: towards a unified theory for learning,” in Advances in Neural Networks—ISNN 2007, vol. 4491 of Lecture Notes in Computer Science, pp. 1094–1103, Springer, Berlin, Germany, 2007.
[9]
H. Isawa, H. Matsushita, and Y. Nishio, “Improved fuzzy adaptive resonance theory combining overlapped category in consideration of connections,” in Proceedings of the IEEE Workshop on Nonlinear Circuit Networks (NCN '08), pp. 8–11, 2008.
[10]
J. He, A.-H. Tan, and C.-L. Tan, “ART-C: a neural architecture for self-organization under constraints,” in Proceedings of the International Joint Conference on Neural Networks (IJCNN '02), pp. 2550–2555, May 2002.
[11]
W. Fung and Y. Liu, “Adaptive categorization of ART networks in robot behavior learning using game-theoretic formulation,” Neural Networks, vol. 16, no. 10, pp. 1403–1420, 2003.
[12]
B. Jerbic, K. Grolinger, and B. Vranjes, “Autonomous agent based on reinforcement learning and adaptive shadowed network,” Artificial Intelligence in Engineering, vol. 13, no. 2, pp. 141–157, 1999.
[13]
B. Moore, “ART 1 and pattern clustering,” in Proceedings of the Connectionist Models Summer School, pp. 174–185, San Mateo, Calif, USA, 1989.
[14]
M. Ghallab, Automated Planning: Theory and Practice, Elsevier/Morgan Kaufmann, Amsterdam, The Netherlands, 2004.
[15]
M. ？vaco, B. ？ekoranja, and B. Jerbi？, “Industrial robotic system with adaptive control,” Procedia Computer Science, vol. 12, pp. 164–169, 2012.
[16]
T. Stipancic, B. Jerbic, and P. Curkovic, “Context-aware system applied in industrial assembly environment,” International Journal of Advanced Robotic Systems, vol. 9, article 234, 2012.
[17]
J. He, A.-H. Tan, C.-L. Tan, and S.-Y. Sung, “On quantitative evaluation of clustering systems,” in Clustering and Information Retrieval, Springer, 2004.
