Fault Tolerance of Oscillatory Neural Network using PMO Oscillator

Oscillatory Neural Networks (ONN) are inevitable when it comes to solving combinatorial optimization problems. This work demonstrates the fault tolerance of the ONN in solving vertex coloring problems in a 4-node network in various configurations at multiple failure levels of the different components of the oscillator. This work validates the network to be extremely robust to failures (limited to 4 nodes), showing tolerance in variations in resistance and capacitances up to 90% and 50% respectively.