The most brutal way to learn about overfitting?
Watching yourself drop hundreds of places when a @kaggle final leaderboard is revealed
-
@benhamner@kaggle overfitting the public LB, which is not the same phenomenon as traditionnal overfitting described in ML books
@chris_bour @benhamner It's the same. Overfitting = selecting a model based on perf on one sample set, being detrimental to generalization.
-
-
@fchollet@chris_bour clever methods of overfitting: http://hunch.net/?p=22Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.