Implementation of Liquid State Machines

Question

Does anybody know of an (open source) implementation of Liquid State Machines?

Answer 1

The Bee simulator is an open-source Spiking Neural Network (SNN) simulator, freely available, specialized in Liquid State Machine (LSM) systems with its core functions fully implemented in C.

                        

It was developed exclusively to solve the specific problems presented by neurorobotics experiments.

• Bee uses the C library pthreads (POSIX threads) in order to speed up the simulation of LSMs by processing input and output in a parallel way. To simplify the user interaction with the software, a Python wrapper is being supplied.

• The neuron model, a special type of model(Leaky_integrate-and-fire) with extra exponential synapses is fixed, following what is presented below, and the solution for the differential equations is calculated by the Euler's method according to the simulation's time step specified by the user.

The simulator automatically generates the liquid(reservoir) in a probabilistic way according to the equation:

For the code in C, you can refer to the following link:https://github.com/ricardodeazambuja/Bee/blob/master/BEE/src/BEE.c

If you want to start learning Artificial Intelligence then go through this video tutorial:

Go through the Artificial Intelligence Course to get clear understanding of artificial intelligence.