2015 AdaptationAlgorithmandTheoryBas

• (Cortes et al., 2015) ⇒ Corinna Cortes, Mehryar Mohri, and Andrés Muñoz Medina. (2015). “Adaptation Algorithm and Theory Based on Generalized Discrepancy.” In: Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2015). ISBN:978-1-4503-3664-2 doi:10.1145/2783258.2783368

Subject Headings:

Cited By

• http://scholar.google.com/scholar?q=%222015%22+Adaptation+Algorithm+and+Theory+Based+on+Generalized+Discrepancy
• http://dl.acm.org/citation.cfm?id=2783258.2783368&preflayout=flat#citedby

Quotes

Author Keywords

• Domain adaptation; learning theory; statistical

Abstract

We present a new algorithm for domain adaptation improving upon the discrepancy minimization algorithm (DM), which was previously shown to outperform a number of popular algorithms designed for this task. Unlike most previous approaches adopted for domain adaptation, our algorithm does not consist of a fixed reweighting of the losses over the training sample. Instead, it uses a reweighting that depends on the hypothesis considered and is based on the minimization of a new measure of generalized discrepancy. We give a detailed description of our algorithm and show that it can be formulated as a convex optimization problem. We also present a detailed theoretical analysis of its learning guarantees, which helps us select its parameters. Finally, we report the results of experiments demonstrating that it improves upon the DM algorithm in several tasks.

References

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2015 AdaptationAlgorithmandTheoryBas	Mehryar Mohri Andrés Muñoz Medina Corinna Cortes			Adaptation Algorithm and Theory Based on Generalized Discrepancy				10.1145/2783258.2783368		2015