Selective Knowledge Transfer for Machine Learning

Selective Knowledge Transfer for Machine Learning

4.11 - 1251 ratings - Source

For selective instance-based transfer, the proposed TransferBoost algorithm uses a novel form of set-based boosting to determine the individual source instances to transfer in learning the target task. TransferBoost reweights instances from each source task based on their collective transferability to the target task, and then performs regular boosting to adjust individual instance weights.In any case, these hindrances disappeared with the addition of slightly more training data. ... The shortcut of always using the average parameter vector works well when all of the source tasks are relevant for transfer to the target task, but this involves ... Additionally, manual selection relies on qualitative (and sometimes incorrect) judgments that the selected tasks will transfer well to the target task.

Title:Selective Knowledge Transfer for Machine Learning
Author:Eric Robert Eaton
Publisher:ProQuest - 2009


You Must CONTINUE and create a free account to access unlimited downloads & streaming