Time Encoding and Decoding (TED) Toolkit


The Time Encoding and Decoding (TED) Toolkit contains instantiations of Time Encoding Machines and Time Decoding Machines. Sample code demonstrating the use of the implementations is included. Note that not all of the algorithms implemented in one language may be available for the other languages listed below.

The Python implementation of the TED toolkit requires scikit-cuda. Interested researchers may wish to read the overview of time encoding and decoding before downloading the toolkits.

Indices of the available encoders based on integrate-and-fire neurons and the associated decoders are available for the Python and MATLAB releases, respectively.

Downloads

MATLAB Release

TED Toolkit 0.7.0 [ code | docs ] (March 21, 2015) ted_matlab_zenodo

Python Release

TED Toolkit 0.7.1 [ code | docs ] (March 21, 2015) ted_python_zenodo

Documentation

The latest MATLAB and Python source code and documentation written by Lev E. Givon, Aurel A. Lazar, Eftychios A. Pnevmatikakis, Yevgeniy B. Slutskiy, Robert J. Turetsky and Yiyin Zhou can be obtained from the Bionet GitHub repository.

  1. Perfect Recovery and Sensitivity Analysis of Time Encoded Bandlimited Signals ,

    Aurel A. Lazar and Laszlo T. Toth , IEEE Transactions on Circuits and Systems-I: Regular Papers , Volume 51 , Number 10 , pp. 2060-2073 , October 2004

  2. Time Encoding with an Integrate-and-Fire Neuron with a Refractory Period ,

    Aurel A. Lazar , Neurocomputing , Volume 58-60 , pp. 53-58 , June 2004

  3. Faithful Representation of Stimuli with a Population of Integrate-and-Fire Neurons ,

    Aurel A. Lazar and Eftychios A. Pnevmatikakis , Neural Computation , Volume 20 , Number 11 , The MIT Press , pp. 2715-2744 , November 2008