ICML 2021 Learning from Biased Data: A Semi-Parametric Approach Spotlight

Learning from Biased Data: A Semi-Parametric Approach

[ Abstract ] [ Visit Learning Theory 6 ] [ Paper ]

[ Paper ]

Abstract: We consider risk minimization problems where the (source) distribution

P_{S}

$P_S$ of the training observations

Z_{1}, \dots, Z_{n}

$Z_1, \ldots, Z_n$ differs from the (target) distribution

P_{T}

$P_T$ involved in the risk that one seeks to minimize. Under the natural assumption that

P_{S}

$P_S$ dominates

P_{T}

$P_T$ , \textit{i.e.} $P_T< \! \!

Chat is not available.