Title :
Application of geometrical algebra to neural computations
Author :
Rodonaia, Irakli ; Rodonaia, Vakhtang
Author_Institution :
Fac. of Comput. Technol. & Eng., Int. Black Sea Univ., Tbilisi, Georgia
Abstract :
Geometrical algebra has been proved to be a powerful mathematical language for neural computations. An overview of some approaches in this area is given in the paper. Applications of the geometrical algebra methods to the geometrical transformations in computer graphics, robot vision and inverse kinematic problem are considered.
Keywords :
algebra; computer graphic equipment; geometry; neural nets; robot kinematics; robot vision; computer graphics; geometrical algebra; geometrical transformations; inverse kinematic problem; mathematical language; neural computations; robot vision; Biological neural networks; Computer graphics; Kinematics; Neurons; Robots; Vectors; computer graphics; geometric product; geometrical algebra; inverse kinematic problem; multivectors; neural computations; robot vision; spinor neuron; spinor representation of rotation;
Conference_Titel :
Application of Information and Communication Technologies (AICT), 2012 6th International Conference on
Conference_Location :
Tbilisi
Print_ISBN :
978-1-4673-1739-9
DOI :
10.1109/ICAICT.2012.6398473
