While implementing retraining on a set cadence is easier, dynamic retraining can prevent models from becoming outdated and optimize computational costs.
Read it!Using Sketch-RNN as a probability estimator to identify the worst sketches of flamingos in 'Quick, Draw!'.
Read it!Prediction intervals are commonly used for linear models but are often underused for random forests. Leveraging the fact that a random forest can provide a conditional distribution instead of just the conditional mean makes prediction intervals relatively straightforward to use in this context.
Read it!