Looks like a solid improvement over what’s being used in the paper. Does it introduce any new optimization difficulties?
I suspect it makes optimisation easier, because we don’t need to compute a tradeoff. But that’s just an informal impression.
Looks like a solid improvement over what’s being used in the paper. Does it introduce any new optimization difficulties?
I suspect it makes optimisation easier, because we don’t need to compute a tradeoff. But that’s just an informal impression.