Timezone: »

Distribution Matching for Transduction
Novi Quadrianto · James Petterson · Alexander Smola

Mon Dec 07 07:00 PM -- 11:59 PM (PST) @

Many transductive inference algorithms assume that distributions over training and test estimates should be related, e.g. by providing a large margin of separation on both sets. We use this idea to design a transduction algorithm which can be used without modification for classification, regression, and structured estimation. At its heart we exploit the fact that for a good learner the distributions over the outputs on training and test sets should match. This is a classical two-sample problem which can be solved efficiently in its most general form by using distance measures in Hilbert Space. It turns out that a number of existing heuristics can be viewed as special cases of our approach.

Author Information

Novi Quadrianto (University of Sussex and HSE)
James Petterson (NICTA)
Alexander Smola (Amazon)

**AWS Machine Learning**

More from the Same Authors