Preprint Article Version 1 NOT YET PEER-REVIEWED

Temporal Modeling of Neural Net Input/Output Behaviors: The Case of XOR

  1. Arizona Center for Integrative Modeling and Simulation (ACIMS), University of Arizona, Tucson, AZ 85721, USA
  2. RTSync Corp. 12500 Park Potomac Ave. #905-S, Potomac, MD 20854, USA
  3. I3S, CNRS, Université Côte d'Azur, France
Version 1 : Received: 27 November 2016 / Approved: 27 November 2016 / Online: 27 November 2016 (05:11:45 CET)

How to cite: Zeigler, B.; Muzy, A. Temporal Modeling of Neural Net Input/Output Behaviors: The Case of XOR. Preprints 2016, 2016110136 (doi: 10.20944/preprints201611.0136.v1). Zeigler, B.; Muzy, A. Temporal Modeling of Neural Net Input/Output Behaviors: The Case of XOR. Preprints 2016, 2016110136 (doi: 10.20944/preprints201611.0136.v1).

Abstract

In the context of modeling and simulation of neural nets, we formulate definitions for behavioral realization of memoryless functions.  The definitions of realization are substantively different for deterministic and stochastic systems constructed of neuron-inspired components. In contrast to Artificial Neural Nets (ANN), and their myriad-layered deep forms, our definitions of realization fundamentally include temporal and probabilistic characteristics of their inputs, state, and outputs. The realizations that we construct, in particular for the XOR logic gate, provide insight into the temporal and probabilistic characteristics that real neural systems might display. We conclude with implications made when contrasting our time-based neural computation systems to ANN for what real brain computations might involve.

Subject Areas

DEVS; neural net computation; XOR function; artificial neural nets; deterministic systems; probabilistic systems

Readers' Comments and Ratings (0)

Discuss and rate this article
Views 51
Downloads 37
Comments 0
Metrics 0
Discuss and rate this article

×
Alerts
Notify me about updates to this article or when a peer-reviewed version is published.