We derive a generalization bound for domain adaptation by using the properties of robust algorithms. Our new bound depends on λ-shift, a measure of prior knowledge regarding the similarity of source and target domain distributions. Based on the generalization bound, we design SVM variants for binary classification and regression domain adaptation algorithms.
|State||Published - 2012|
|Event||International Symposium on Artificial Intelligence and Mathematics, ISAIM 2012 - Fort Lauderdale, FL, United States|
Duration: 9 Jan 2012 → 11 Jan 2012
|Conference||International Symposium on Artificial Intelligence and Mathematics, ISAIM 2012|
|City||Fort Lauderdale, FL|
|Period||9/01/12 → 11/01/12|