Reference for learners

Glossary

• Constraint optimization: A class of optimization problems where an optimal solution needs to be found while taking into account certain parameters/constraints in how it should look like. Examples would be the map coloring problem (no neighboring cells can share a color), a Sudoku puzzle (a number cannot repeat in a given cell, column or row), or a mechanical problem such as inverse kinematics.

• LIF: Leaky Integrate and Fire neuron. The basic spiking neuron model utilized in neuromorphic computing.

• Map coloring problem: Classic optimization problem of coloring a map (in the abstract sense, neighboring cells of arbitrary shape) so no 2 neighboring locations share a “color”. Can be easily transformed into a graph optimization problem, and from there derive a neuromorphic implementation. The 2D solution is trivial nowadays, but higher dimensional configurations are not so simple.

• ML/AI: The Machine Learning/ Artifical Intelligence family of algorithms and methods.

• Neuromorphic Computing (NMC or NC): Biologically inspired branch of computing whose main “computational unit” is the spiking neuron, and the networks built from combining up to tens or hundreds of thousands of these neurons.

Reading materials for further learning

The Open Neuromorphic community is a good starting point for an overview of the field. Otherwise each section presents its own suggestion for further reading material.