A Temporal Bias Correction using a Machine Learning Attention model
Stats
気温最大値に関する20年間(1989年〜2008年)のRMSE:東京で2.57、アブジャで1.88
Quotes
"Climate models are biased with respect to real world observations and usually need to be calibrated prior to impact studies."
"Our model opens up the opportunity to do BC temporally."