Adaptation Algorithm and Theory Based on Generalized Discrepancy

TL;DR

This paper introduces a novel domain adaptation algorithm based on generalized discrepancy, offering improved theoretical guarantees and empirical performance over previous discrepancy minimization methods.

Contribution

The paper proposes a new adaptation algorithm with a solid theoretical foundation and more favorable learning bounds, differing from fixed reweighting approaches.

Findings

Our algorithm outperforms discrepancy minimization in experiments.

It benefits from better theoretical learning bounds.

The method provides efficient solutions for the optimization problem.

Abstract

We present a new algorithm for domain adaptation improving upon a discrepancy minimization algorithm previously shown to outperform a number of algorithms for this task. Unlike many previous algorithms for domain adaptation, our algorithm does not consist of a fixed reweighting of the losses over the training sample. We show that our algorithm benefits from a solid theoretical foundation and more favorable learning bounds than discrepancy minimization. We present a detailed description of our algorithm and give several efficient solutions for solving its optimization problem. We also report the results of several experiments showing that it outperforms discrepancy minimization.