Re:Ask HN: Is there a case for small / medium data
by shoo on Friday May 28, 2021 @03:20AM
(#27312599)
In some cases there can be a large improvement going from status quo (if status quo is rather lacklustre) to a simple model, and it may not be worth doing anything more complicated if the accuracy of the component in question is no longer a bottleneck of overall system performance.
Maybe a simple model with a well-chosen prior informed by domain knowledge does the job.
Maybe a simple model with a well-chosen prior informed by domain knowledge does the job.