Domain Generalization with Adversarial Feature Learning

By Haoliang Li, Sinno Jialin Pan, Shiqi Wang, Alex ChiChung Kot. 2018

1,510 citations recorded by the source corpus.

In this paper, we tackle the problem of domain generalization: how to learn a generalized feature representation for an "unseen" target domain by taking the advantage of multiple seen source-domain data. We present a novel framework based on adversarial autoencoders to learn a generalized latent feature representation across domains for domain generalization. To be specific, we extend adversarial autoencoders by imposing the Maximum Mean Discrepancy (MMD) measure to align the distributions among different domains, and matching the aligned distribution to an arbitrary prior distribution via adversarial feature learning. In this way, the learned feature representation is supposed to be universal to the seen source domains because of the MMD regularization, and is expected to generalize well on the target domain because of the introduction of the prior distribution. We proposed an algorithm to jointly train different components of our proposed framework. Extensive experiments on various vision tasks demonstrate that our proposed framework can learn better generalized features for the unseen target domain compared with state-of-the-art domain generalization methods.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Learning Robust Global Representations by Penalizing Local Predictive Power (2019, methodology): “…invariance via adversarial learning (or further extensions) (Ganin et al., 2016; Bousmalis et al., 2017; Tzeng et al., 2017; Xie et al., 2018; Hoffman et al., 2018; Long et al., 2018; Zhao et al., 2018b; Kumar et al., 2018; Li et al., 2018b; Zhao et al., 2018a; Schoenauer-Sebag et al., 2019).”
GAN Inversion: A Survey (2021, methodology): “[175] use adversarial autoencoder [181] to learn invariant features for domain generalization.”
Learning Robust Representations by Projecting Superficial Statistics Out (2018, background): “s mostly fall into two categories: reducing the inter-domain differences of representations through adversarial (or similar) techniques (Ghifary et al., 2015; Wang et al., 2016; Motiian et al., 2017; Li et al., 2018; Carlucci et al., 2018), or building an ensemble of one-for-each-domain deep 2 Publi”
In Search of Lost Domain Generalization (2020, background): “– MMD [Li et al., 2018b] matches the MMD [Gretton et al., 2012] of feature distributions.”
Domain Generalization with MixStyle (2021, methodology): “Recently, meta-learning has drawn increasing attention from the DG community (Li et al., 2018a; Balaji et al., 2018; Dou et al., 2019).”
Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data (2019, background): “[28] imposed Maximum Mean Discrepancy measure to align the distributions among different domains and train the network with adversarial feature learning.”
Agnostic Domain Generalization (2018, methodology): “Both the works [12, 24] exploit deep learning autoencoders for domain generalization still focusing on representation learning.”
Cross-Domain Face Presentation Attack Detection via Multi-Domain Disentangled Representation Learning (2020, methodology): “However, our approach performs better than MADDG [38] and MMD-AAE [26] multi-domain disentangled representation learning.”
Towards Out-Of-Distribution Generalization: A Survey (2021, background): “Domain generalization methods [6], [7], [8], [9] mainly focus on real scenarios and utilize data collected from different domains.”
Open Compound Domain Adaptation (2019, background): “Domain generalization [52, 23, 22] and domain agnostic learning [39, 5] aim to learn universal representations that can be applied in a domain-invariant manner.”
Domain Generalization via Model-Agnostic Learning of Semantic Features (2019, background): “A promising direction is to extract task-speciﬁc but domain-invariant features [12, 28, 30, 34, 35].”
Heterogeneous Domain Generalization Via Domain Mixup (2020, background): “Domain generalization (DG) [8, 9, 10, 11, 12, 13] aims to address the harder setting where the target domain is not available during training stage.”
Cross-Domain First Person Audio-Visual Action Recognition through Relative Norm Alignment (2021, methodology): “Previous approaches in DG are mostly designed for image data [6, 56, 30, 17, 31, 5] and are divided in feature-based and data-based methods.”
Domain Generalization: A Survey (2021, methodology): “Action recognition - IXMAS [68] 1,650 5 5 camera views, 10 subjects (see [31]) - UCF-HMDB [69], [70] 3,809 2 Data sources (2 datasets) (see [71])”
Measuring Domain Shift for Deep Learning in Histopathology (2020, background): “Domain generalization: Previous work in domain generalization often try to learn uniﬁed representations from multiple source domains [26]–[28].”
Domain-agnostic Question-Answering with Adversarial Training (2019, methodology): “We apply meta learning to domain generalization (Li et al., 2018a, 2019; Balaji et al., 2018) to simulate train/test domain shift.”
Reversible Instance Normalization for Accurate Time-Series Forecasting against Distribution Shift (2022, methodology): “Domain adaptation (Tzeng et al., 2017; Ganin et al., 2016; Wang et al., 2018) and domain generalization (Wang et al., 2021; Li et al., 2018; Muandet et al., 2013) are common ways to alleviate the distribution shift.”