So I guess we don't necessarily have to estimate the density distribution itself. We can just use the discrepancy as a metric to pick samples.
Could this be mitigated by starting out with a regular grid then uniformally randomly perturbing each point some way?
So I guess we don't necessarily have to estimate the density distribution itself. We can just use the discrepancy as a metric to pick samples.
Could this be mitigated by starting out with a regular grid then uniformally randomly perturbing each point some way?