A Discriminative Technique for Multiple-Source Adaptation
Authors:
- Corinna CortesCCCorinna Cortes
- Mehryar MohriMMMehryar Mohri
- Ananda Theertha SureshASAnanda Theertha Suresh
- Ningshan ZhangNZNingshan Zhang
Preprint
Abstract
We present a new discriminative technique for the multiple-source adaptation, MSA, problem. Unlike previous work, which relies on density estimation for each source domain, our solution only requires conditional probabilities that can easily be accurately estimated from unlabeled data from the source domains. We give a detailed analysis of our new technique, including general guarantees based on R\'enyi divergences, and learning bounds when conditional Maxent is used for estimating conditional probabilities for a point to belong to a source domain. We show that these guarantees compare favorably to those that can be derived for the generative solution, using kernel density estimation. Our experiments with real-world applications further demonstrate that our new discriminative MSA algorithm outperforms the previous generative solution as well as other domain adaptation baselines.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Multi-Source Domain Adaptation: A Causal ViewProceedings of the AAAI Conference on Artificial Intelligence, 2015
- Exploiting Low-Rank Structure from Latent Domains for Domain Generalization2014
- Speaker adaptation of context dependent deep neural networks2013
- Discovering Latent Domains for Multisource Domain Adaptation2012
- Unbiased look at dataset bias2011
- Confidence-weighted linear classification2008
- Adaptive Mixtures of Local ExpertsNeural Computation, 1991