Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation

Abstract

Domain adaptation (DA) is the topical problem of adapting models from labelled source datasets so that they perform well on target datasets where only unlabelled or partially labelled data is available. Many methods have been proposed to address this problem through different ways to minimise the domain shift between source and target datasets. In this paper we take an orthogonal perspective and propose a framework to further enhance performance by meta-learning the initial conditions of existing DA algorithms. This is challenging compared to the more widely considered setting of few-shot meta-learning, due to the length of the computation graph involved. Therefore we propose an online shortest-path meta-learning framework that is both computationally tractable and practically effective for improving DA performance. We present variants for both multi-source unsupervised domain adaptation (MSDA), and semi-supervised domain adaptation (SSDA). Importantly, our approach is agnostic to the base adaptation algorithm, and can be applied to improve many techniques. Experimentally, we demonstrate improvements on classic (DANN) and recent (MCD and MME) techniques for MSDA and SSDA, and ultimately achieve state of the art results on several DA benchmarks including the largest scale DomainNet.

arXiv commentECCV 2020 CR version

References in corpus (18)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Learning Transferable Features with Deep Adaptation Networks
• Deep Domain Confusion: Maximizing for Domain Invariance
• Conditional Adversarial Domain Adaptation
• Frustratingly Easy Domain Adaptation
• DARTS: Differentiable Architecture Search
• Theoretical Models of Learning to Learn
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Domain Separation Networks
• On First-Order Meta-Learning Algorithms
• Domain Generalization via Model-Agnostic Learning of Semantic Features
• Meta-Learning with Implicit Gradients
• Meta-Gradient Reinforcement Learning
• Online Meta-Learning
• Concurrent Meta Reinforcement Learning
• Wasserstein Distance Based Domain Adaptation for Object Detection
• Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation

Cited by in corpus (2)

• FixBi: Bridging Domain Spaces for Unsupervised Domain Adaptation
• Dynamic Transfer for Multi-Source Domain Adaptation