3 ms·
Interesting. What you are describing reminds me of Low discrepancy sequences: https://en.wikipedia.org/wiki/Low-discrepancy_sequence https://en.wikipedia.org/w
by eru 2y ago
Interesting.
What you are describing reminds me of Low discrepancy sequences: https://en.wikipedia.org/wiki/Low-discrepancy_sequence https://en.wikipedia.org/wiki/Low-discrepancy_sequence
Though these methods have their problems and blind-spots, too, and are often outdone by random sampling with even slightly higher sample count, while preserving all the simplicity and (statistical) guarantees you get from randomness.