We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.DS

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Dynamical Systems

Title: Permutation Binary Neural Networks: Analysis of Periodic Orbits and Its Applications

Abstract: This paper presents a permutation binary neural network characterized by local binary connection, global permutation connection, and the signum activation function. The dynamics is described by a difference equation of binary state variables. Depending on the connection, the network generates various periodic orbits of binary vectors. The binary/permutation connection brings benefits to precise analysis and to FPGA based hardware implementation. In order to consider the periodic orbits, we introduce three tools: a composition return map for visualization of the dynamics, two feature quantities for classification of periodic orbits, and an FPGA based hardware prototype for engineering applications. Using the tools, we have analyzed all the 6-dimensional networks. Typical periodic orbits are confirmed experimentally.
Comments: 16 pages, 10 Postscript figures, uses aims.cls
Subjects: Dynamical Systems (math.DS); Chaotic Dynamics (nlin.CD)
MSC classes: 37B15, 68T07
Cite as: arXiv:2201.09219 [math.DS]
  (or arXiv:2201.09219v1 [math.DS] for this version)

Submission history

From: Toshimichi Saito [view email]
[v1] Sun, 23 Jan 2022 10:01:06 GMT (2566kb)

Link back to: arXiv, form interface, contact.