Draft:Lottery ticket hypothesis

In machine learning, the lottery ticket hypothesis is that artificial neural networks with random weights can contain subnetworks which entirely by chance can be tuned to a similar level of performance as the complete network.

Malach et. al. have proved a stronger version of the hypothesis, which is that a sufficiently overparameterized untuned network will typically contain a subnetwork that is already an approximation to the given goal, even before tuning. A similar result has been proven for the special case of convolutional neural networks.