نتایج جستجو برای: underestimation

تعداد نتایج: 6080  

Journal: :Communications in computer and information science 2021

It is fair to say that many of the prominent examples bias in Machine Learning (ML) arise from training data. In fact, some would argue supervised ML algorithms cannot be biased, they reflect data on which are trained. this paper, we demonstrate how can misrepresent through underestimation. We show irreducible error, regularization, and feature class imbalance contribute The paper concludes wit...

نمودار تعداد نتایج جستجو در هر سال

با کلیک روی نمودار نتایج را به سال انتشار فیلتر کنید