Title :
Non-Hermitian associative memories spontaneously generating dynamical attractors
Author_Institution :
Res. Center for Adv. Sci. & Technol., Tokyo Univ., Japan
Abstract :
Experimental results on complex-valued associative memories are reported. When the memory has some dynamical attractors, the weighting matrix should generally be non-Hermitian. The non-Hermitian associative memories have an advantage that they can express trajectories inherently smoother than conventional real-number memories. In this paper, spontaneous variations of simple stationary attractors into dynamical and complicated attractors are reported when the construction of weighting matrix changes from the Hermitian to the non-Hermitian.
Keywords :
content-addressable storage; matrix algebra; neural nets; complex-valued associative memories; dynamical attractors generation; nonHermitian associative memories; spontaneous variations; stationary attractors; weighting matrix; Associative memory; Biological system modeling; Eigenvalues and eigenfunctions; Evolution (biology); Feedback loop; Ice; Motion control; Neural networks; Neurons; Vectors;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714248