If a simulation is trying to predict the behavior of a simple mechanical system, it's probably accurate. If it's trying to predict the behavior of a biological system, or any other complex system, it's probably not.
-
-
This Tweet is unavailable.
-
Getting my guess of fluid dynamics in before John gives the answer
End of conversation
-
-
-
Doesn't well running machine learning make accurate predictions all the time? FB is very good at predicting what ad someone will click, even though that's quite a complex system.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
He must mean more than a tautology...
End of conversation
-
-
-
- Instead of relying on a single parameterization, rely on a diverse set of model ensembles - Robustly and rigorously cross validate your model, making sure to not leak data across folds It turns out that almost everyone underestimates the uncertainty of their forecasts.
-
This is almost of a human problem as a modeling problem: people prefer optimistic forecasts that drastically understate uncertainties to realistic forecasts with robust confidence intervals because they don’t feel as actionable.
End of conversation
New conversation -
-
-
Weather prediction has become increasingly accurate over time. Which is kinda the Ur complex system.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Hamming's "Art of science and engineering" has a chapter on reliability of simulation. Said that for complex systems the best you can do is the method of scenarios. He'd fend off people wanting ecological sims by asking for laws and data up to his standard.
-
(I don't know if he's right, but he certainly had experience.)
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.