22,647 Hits in 1.6 sec

Domain Invariant Adversarial Learning [article]

Matan Levi, Idan Attias, Aryeh Kontorovich
2021 arXiv   pre-print
We present a new adversarial training method, Domain Invariant Adversarial Learning (DIAL), which learns a feature representation that is both robust and domain invariant.  ...  In the case where the source domain consists of natural examples and the target domain is the adversarially perturbed examples, our method learns a feature representation constrained not to discriminate  ...  DOMAIN INVARIANT ADVERSARIAL LEARNING APPROACH In this section, we introduce our Domain Invariant Adversarial Learning (DIAL) approach for adversarial training.  ... 
arXiv:2104.00322v3 fatcat:d3xirohln5durorico7y57q7j4

Adversarial Invariant Feature Learning with Accuracy Constraint for Domain Generalization [article]

Kei Akuzawa, Yusuke Iwasawa, Yutaka Matsuo
2020 arXiv   pre-print
We then propose a novel method adversarial feature learning with accuracy constraint (AFLAC), which explicitly leads to that invariance on adversarial training.  ...  Learning domain-invariant representation is a dominant approach for domain generalization (DG), where we need to build a classifier that is robust toward domain shifts.  ...  After deriving the theorems, we propose a novel method adversarial feature learning with accuracy constraint (AFLAC), which leads to that invariance on adversarial training.  ... 
arXiv:1904.12543v3 fatcat:mvw7bo7nx5bbtgjhb5ankqvuva

Domain-Invariant Adversarial Learning for Unsupervised Domain Adaption [article]

Yexun Zhang, Ya Zhang, Yanfeng Wang, Qi Tian
2018 arXiv   pre-print
The same encoder is shared between the source and target domains which is expected to extract domain-invariant representations with the help of an adversarial discriminator.  ...  In this paper, we thus propose a simple but effective model for unsupervised domain adaption leveraging adversarial learning.  ...  Conclusion In this paper, we propose a Domain-Invariant Adversarial Learning (DIAL) network for unsupervised domain adaption, which is shown to extract both domain-invariant and discriminative features  ... 
arXiv:1811.12751v1 fatcat:quexyypeovbvndgu4wzrlutt2u

Attentive Adversarial Learning for Domain-invariant Training

Zhong Meng, Jinyu Li, Yifan Gong
2019 ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
More generally, the same methodology can improve any adversarial learning system with an auxiliary discriminator.  ...  Adversarial domain-invariant training (ADIT) proves to be effective in suppressing the effects of domain variability in acoustic modeling and has led to improved performance in automatic speech recognition  ...  To achieve domain-robustness, one solution is to learn a domain-invariant and senone-discriminative deep hidden feature in the DNN acoustic model through adversarial multi-task learning and make senone  ... 
doi:10.1109/icassp.2019.8683486 dblp:conf/icassp/MengLG19a fatcat:eb2rsxeyivfstoz5ux2xt5al3e

QAGAN: Adversarial Approach To Learning Domain Invariant Language Features [article]

Shubham Shrivastava, Kaiyue Wang
2022 arXiv   pre-print
In this paper, we explore adversarial training approach towards learning domain-invariant features so that language models can generalize well to out-of-domain datasets.  ...  the model to learn domain invariant embedding and bring them closer in the multi-dimensional space.  ...  Since then, learning domain invariant features through adversarial training framework has gain increasing attention in research community.  ... 
arXiv:2206.12388v1 fatcat:pzm24afokfhgbnhapoblbbbebm

Adversarial Learning of Raw Speech Features for Domain Invariant Speech Recognition [article]

Aditay Tripathi, Aanchan Mohan, Saket Anand, Maneesh Singh
2018 arXiv   pre-print
Promising empirical results indicate the strength of adversarial training for unsupervised domain adaptation in ASR, thereby emphasizing the ability of DANNs to learn domain invariant features from raw  ...  This paper explores the application of adversarial training to learn features from raw speech that are invariant to acoustic variability.  ...  Section 2 reviews related work done in the area of domain-invariant feature learning using domain adversarial neural networks.  ... 
arXiv:1805.08615v1 fatcat:ehbdqulelfdgbk4yqtlfitvbau

Generalizing to unseen domains via distribution matching [article]

Isabela Albuquerque, João Monteiro, Mohammad Darvishi, Tiago H. Falk, Ioannis Mitliagkas
2021 arXiv   pre-print
Minimizing the terms in our bound yields an adversarial formulation which estimates and minimizes pairwise discrepancies.  ...  Supervised learning results typically rely on assumptions of i.i.d. data. Unfortunately, those assumptions are commonly violated in practice.  ...  to learn robust representations [8, 9] , and learning domain-invariant representations [10] , among other approaches.  ... 
arXiv:1911.00804v6 fatcat:5tbbgs7ee5dc7iji625babygfq

Domain-invariant adversarial learning with conditional distribution alignment for unsupervised domain adaptation

Xingmei Wang, Boxuan Sun, Hongbin Dong
2020 IET Computer Vision  
The marginal distribution is aligned in the adversarial learning process of extracting domain-invariant features.  ...  In this study, domain-invariant adversarial learning with conditional distribution alignment is proposed to alleviate the effect of domain shift with label shift.  ...  Domain-invariant adversarial learning In a traditional auto-encoder, there is only a reconstruction structure.  ... 
doi:10.1049/iet-cvi.2019.0514 fatcat:q5pniqk44rhvbnt2njlwqwveii

Representation via Representations: Domain Generalization via Adversarially Learned Invariant Representations [article]

Zhun Deng, Frances Ding, Cynthia Dwork, Rachel Hong, Giovanni Parmigiani, Prasad Patil, Pragya Sur
2020 arXiv   pre-print
We examine adversarial censoring techniques for learning invariant representations from multiple "studies" (or domains), where each study is drawn according to a distribution on domains.  ...  To our knowledge, our results provide the first formal guarantees of these kinds for adversarial invariant domain generalization.  ...  To this end, we adopt an adversarial censored learning approach.  ... 
arXiv:2006.11478v1 fatcat:djprbhsabvforbl7iecyuslvye

Towards Category and Domain Alignment: Category-Invariant Feature Enhancement for Adversarial Domain Adaptation [article]

Yuan Wu, Diana Inkpen, Ahmed El-Roby
2021 arXiv   pre-print
In order to alleviate these issues, we propose category-invariant feature enhancement (CIFE), a general mechanism that enhances the adversarial domain adaptation through optimizing the adaptability.  ...  be domain-invariant needs to sacrifice domain-specific variations, resulting in weaker discriminability.  ...  Adversarial Domain Adaptation The key idea of adversarial domain adaptation is to learn domain-invariant features that can be generalized across domains.  ... 
arXiv:2108.06583v1 fatcat:pvbwmfxm2zf3jgshpgho6cjkfi

Adversarial Multi-Task Learning of Deep Neural Networks for Robust Speech Recognition

Yusuke Shinohara
2016 Interspeech 2016  
In this paper, we propose adversarial multi-task learning of DNNs for explicitly enhancing the invariance of representations.  ...  What is different from the standard multi-task learning is that the representation is learned adversarially to the secondary task, so that representation with low domain-classification accuracy is induced  ...  [15] [16] proposed to use adversarial framework for learning domain-invariant representations, evaluated their algorithm in unsupervised domain adaptation tasks, and achieved state-of-the-art performance  ... 
doi:10.21437/interspeech.2016-879 dblp:conf/interspeech/Shinohara16a fatcat:p7hrt75grjcx7cwvaaagywhqom

Domain-agnostic Question-Answering with Adversarial Training [article]

Seanie Lee, Donggyu Kim, Jangwon Park
2019 arXiv   pre-print
The training is performed in the adversarial manner, where the two models constantly compete, so that QA model can learn domain-invariant features.  ...  Adapting models to new domain without finetuning is a challenging problem in deep learning.  ...  Each training dataset is considered different domain for adversarial learning in which QA model learns domain-invariant feature representation by competing with discriminator.  ... 
arXiv:1910.09342v2 fatcat:5khf7io3nfa5lcuyu2vk762iyu

TSAL: Two steps Adversarial learning based domain adaptation

Haidi Hasan Badr, Nayer Mahmoud Wanas, Magda Fayek
2021 Proceedings of the ... International Florida Artificial Intelligence Research Society Conference  
TSAL utilizes joint adversarial learning with class information and domain alignment deep network architecture to learn both domain-invariant and domain-specific features extractors.  ...  Adversarial learning is a strategy for learning domain-transferable features in robust deep networks. This paper introduces the TSAL paradigm, a two-step adversarial learning framework.  ...  Chen and Cardie (Chen and Cardie 2018) propose a multinomial adversarial network (MAN) that learns invariant features across multiple domains.  ... 
doi:10.32473/flairs.v34i1.128510 fatcat:6qpuq4tqlvhvbpfy3cai42rceu

Dual Adversarial Co-Learning for Multi-Domain Text Classification [article]

Yuan Wu, Yuhong Guo
2019 arXiv   pre-print
In this paper we propose a novel dual adversarial co-learning approach for multi-domain text classification (MDTC).  ...  The approach learns shared-private networks for feature extraction and deploys dual adversarial regularizations to align features across different domains and between labeled and unlabeled data simultaneously  ...  in the target domain, aiming to help learn domain-invariant features.  ... 
arXiv:1909.08203v1 fatcat:rrnw2zpkqbeaxlatfhpvjerhg4

Zero-Shot Dense Retrieval with Momentum Adversarial Domain Invariant Representations [article]

Ji Xin, Chenyan Xiong, Ashwin Srinivasan, Ankita Sharma, Damien Jose, Paul N. Bennett
2021 arXiv   pre-print
source versus target, and then adversarially updates the DR encoder to learn domain invariant representations.  ...  To achieve that, we propose Momentum adversarial Domain Invariant Representation learning (MoDIR), which introduces a momentum method in the DR training process to train a domain classifier distinguishing  ...  In this paper, we present Momentum Adversarial Domain Invariant Representations learning (MoDIR), to improve the generalization ability of zero-shot dense retrieval (ZeroDR).  ... 
arXiv:2110.07581v1 fatcat:ss4edwkvprgeff7mvz4qhoes2e
« Previous Showing results 1 — 15 out of 22,647 results