Unsupervised Domain Adaptation With Class-Aware Memory Alignment

doi:10.1109/TNNLS.2023.3238063

Unsupervised Domain Adaptation With Class-Aware Memory Alignment

2024-07-01

2

PRE

AI

DOI：10.1109/TNNLS.2023.3238063

Original

Original request for help

Save

Abstract

En 中文

Unsupervised domain adaptation (UDA) is to make predictions on unlabeled target domain by learning the knowledge from a label-rich source domain. In practice, existing UDA approaches mainly focus on minimizing the discrepancy between different domains by mini-batch training, where only a few instances are accessible at each iteration. Due to the randomness of sampling, such a batch-level alignment pattern is unstable and may lead to misalignment. To alleviate this risk, we propose class-aware memory alignment (CMA) that models the distributions of the two domains by two auxiliary class-aware memories and performs domain adaptation on these predefined memories. CMA is designed with two distinct characteristics: class-aware memories that create two symmetrical class-aware distributions for different domains and two reliability-based filtering strategies that enhance the reliability of the constructed memory. We further design a unified memory-based loss to jointly improve the transferability and discriminability of features in the memories. State-of-the-art (SOTA) comparisons and careful ablation studies show the effectiveness of our proposed CMA.

Keywords:

Adaptation models

Filtering

Training

Feature extraction

Task analysis

Reliability engineering

Measurement

Classification

domain adaptation

memory-based alignment

reliability-based filtering

Unsupervised Domain Adaptation With Class-Aware Memory Alignment

Abstract

Journal

Organization