8,433 Hits in 4.4 sec

Continual Unsupervised Domain Adaptation for Semantic Segmentation [article]

Joonhyuk Kim, Sahng-Min Yoo, Gyeong-Moon Park, Jong-Hwan Kim
2021 arXiv   pre-print
Moreover, Continual UDA, which deals with more practical scenarios with multiple target domains in the continual learning setting, has not been actively explored.  ...  Unsupervised Domain Adaptation (UDA) for semantic segmentation has been favorably applied to real-world scenarios in which pixel-level labels are hard to be obtained.  ...  Therefore, many Unsupervised Domain Adaptation (UDA) techniques [14, 32, 35, 4, 44, 37] , which aim to adapt the network trained on synthetic images to real images, have been introduced to solve the domain  ... 
arXiv:2010.09236v2 fatcat:zou42auycfd7zihjb3qch2wgv4

RDA: Robust Domain Adaptation via Fourier Adversarial Attacking [article]

Jiaxing Huang, Dayan Guan, Aoran Xiao, Shijian Lu
2021 arXiv   pre-print
With FAA-generated samples, the training can continue the 'random walk' and drift into an area with a flat loss landscape, leading to more robust domain adaptation.  ...  Unsupervised domain adaptation (UDA) involves a supervised loss in a labeled source domain and an unsupervised loss in an unlabeled target domain, which often faces more severe overfitting (than classical  ...  Figure 1 . 1 Our robust domain adaptation alleviates overfitting effectively: Both supervised learning with source data (row 1) and unsupervised learning with target data (in rows 2 and 3 for adversarial  ... 
arXiv:2106.02874v3 fatcat:w5gzhaxuprbrtesxmrxfrilwm4

Unsupervised Domain Adaptation for Dialogue Sequence Labeling Based on Hierarchical Adversarial Training

Shota Orihashi, Mana Ihori, Tomohiro Tanaka, Ryo Masumura
2020 Interspeech 2020  
Index Terms: dialogue sequence labeling, unsupervised domain adaptation, hierarchical adversarial training 2. Related work 2.1.  ...  This paper presents a novel unsupervised domain adaptation method for dialogue sequence labeling.  ...  Unsupervised domain adaptation Unsupervised domain adaptation is the technique to convert a machine learning model from a source domain into the target domain equivalent by using unlabeled data of the  ... 
doi:10.21437/interspeech.2020-2010 dblp:conf/interspeech/OrihashiITM20 fatcat:xlkbbfo7jnf77eripy42h65hgu

Consistency Regularization with High-dimensional Non-adversarial Source-guided Perturbation for Unsupervised Domain Adaptation in Segmentation [article]

Kaihong Wang, Chenhongyi Yang, Margrit Betke
2020 arXiv   pre-print
Unsupervised domain adaptation for semantic segmentation has been intensively studied due to the low cost of the pixel-level annotation for synthetic data.  ...  with annotations from the latter.  ...  We also show that domain adaptation is achievable in a bidirectional way through a continuous parameterization of the two domains, without requiring adversarial training; 2.  ... 
arXiv:2009.08610v1 fatcat:ftslqpxe6bf45gpisrq6cbv5hu

Mining Label Distribution Drift in Unsupervised Domain Adaptation [article]

Peizhao Li, Zhengming Ding, Hongfu Liu
2020 arXiv   pre-print
Unsupervised domain adaptation targets to transfer task knowledge from labeled source domain to related yet unlabeled target domain, and is catching extensive interests from academic and industrial areas  ...  Next, we propose Label distribution Matching Domain Adversarial Network (LMDAN) to handle data distribution shift and label distribution drift jointly.  ...  Further, by the success of Generative Adversarial Network [15] , adversarial learning in deep models for unsupervised domain adaptation are continually deliver favorable performance [22, 5, 34] .  ... 
arXiv:2006.09565v1 fatcat:pkpzkxeiijhqfas3h23zjh5rcy

Dual Mixup Regularized Learning for Adversarial Domain Adaptation [article]

Yuan Wu, Diana Inkpen, Ahmed El-Roby
2020 arXiv   pre-print
Recent advances on unsupervised domain adaptation (UDA) rely on adversarial learning to disentangle the explanatory and transferable features for domain adaptation.  ...  However, there are two issues with the existing methods.  ...  More recently, unsupervised domain adaptation methods are largely focusing on learning domain-invariant features by using adversarial training [4] .  ... 
arXiv:2007.03141v2 fatcat:fdmh32xdijerdiopgsbls3qvqi

Adversarial Learning based Discriminative Domain Adaptation for Geospatial Image Analysis

Nikhil Makkar, Hsiuhan Lexie Yang, Saurabh Prasad
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
We are using adversarial learning to extract discriminative target domain features that are aligned with source domain.  ...  In this work, we use adversarial learning for domain adaptation for remote sensing applications.  ...  METHODOLOGY This work is motivated by discriminative domain adaptation framework for unsupervised domain adaptation using adversar- ial learning, proposed in [9] as Adversarial Discriminative Domain  ... 
doi:10.1109/jstars.2021.3132259 fatcat:5ppi25cwirc2bmnlgolauiwga4

A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation [article]

Alexander H. Liu, Yen-Cheng Liu, Yu-Ying Yeh, Yu-Chiang Frank Wang
2018 arXiv   pre-print
Realized by adversarial training with additional ability to exploit domain-specific information, the proposed network is able to perform continuous cross-domain image translation and manipulation, and  ...  In addition, the resulting feature representation exhibits superior performance of unsupervised domain adaptation, which also verifies the effectiveness of the proposed model in learning disentangled features  ...  Unsupervised domain adaption with domain-invariant representation Unsupervised domain adaption (UDA) aims to classify samples in target domain while labels are only available in the source domain.  ... 
arXiv:1809.01361v3 fatcat:mohw4eefg5ajrotb6dlivsr7du

Unsupervised domain adaptation for cross-modality liver segmentation via joint adversarial learning and self-learning [article]

Jin Hong, Simon Chun-Ho Yu, Weitian Chen
2022 arXiv   pre-print
In this work, we report a novel unsupervised domain adaptation framework for cross-modality liver segmentation via joint adversarial learning and self-learning.  ...  Using the public data sets, our experiments demonstrated the proposed unsupervised domain adaptation framework reached four supervised learning methods with a Dice score 0.912 plus or minus 0.037 (mean  ...  Thus, we name it as shape-entropy-aware adversarial learning for unsupervised domain adaptation.  ... 
arXiv:2109.05664v3 fatcat:htdtrwkb6vci5orzardyor33me

Augmented Cyclic Adversarial Learning for Low Resource Domain Adaptation [article]

Ehsan Hosseini-Asl, Yingbo Zhou, Caiming Xiong, Richard Socher
2019 arXiv   pre-print
Domain adaptation deals with the challenge of adapting a model trained from a data-rich source domain to perform well in a data-poor target domain.  ...  CycleGAN is a powerful framework that efficiently learns to map inputs from one domain to another using adversarial training and a cycle-consistency constraint.  ...  Tzeng et al. (2017) proposes adversarial discriminative domain adaptation (ADDA), where adversarial learning is employed to match the representation learned from the source and target domain.  ... 
arXiv:1807.00374v4 fatcat:4zxqmd2kv5erbiwdreycxojowq

Incremental Adversarial Domain Adaptation for Continually Changing Environments [article]

Markus Wulfmeier, Alex Bewley, Ingmar Posner
2018 arXiv   pre-print
While unsupervised domain adaptation aims to address this challenge, current approaches do not utilise the continuity of the occurring shifts.  ...  Our work presents an adversarial approach for lifelong, incremental domain adaptation which benefits from unsupervised alignment to a series of intermediate domains which successively diverge from the  ...  The approach extends adversarial domain adaptation approaches [3] aiming to facilitate learning a feature encoding f which is invariant with respect to the origin domain of its input data.  ... 
arXiv:1712.07436v2 fatcat:g47t72heljffbgmwulw6wmqk6m

AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation [article]

Jogendra Nath Kundu, Phani Krishna Uppala, Anuj Pahuja, R. Venkatesh Babu
2018 arXiv   pre-print
Recent adversarial approaches for domain adaption have performed well in mitigating the differences between the source and target domains.  ...  In this work, we propose AdaDepth - an unsupervised domain adaptation strategy for the pixel-wise regression task of monocular depth estimation.  ...  Domain Consistency Regularization (DCR) Since we start the adversarial learning after training on synthetic images, the resultant adaptation via adversarial objective should not distort the rich learned  ... 
arXiv:1803.01599v2 fatcat:gad7mrdeznenppfg4ozjlwnyeu

AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation

Jogendra Nath Kundu, Phani Krishna Uppala, Anuj Pahuja, R. Venkatesh Babu
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Recent adversarial approaches for domain adaption have performed well in mitigating the differences between the source and target domains.  ...  The proposed approach is devoid of above limitations through a) adversarial learning and b) explicit imposition of content consistency on the adapted target representation.  ...  Domain Consistency Regularization (DCR) Since we start the adversarial learning after training on synthetic images, the resultant adaptation via adversarial objective should not distort the rich learned  ... 
doi:10.1109/cvpr.2018.00281 dblp:conf/cvpr/KunduUPB18 fatcat:wdistqaqhzfpbdm5n7crlivowq

When Autonomous Systems Meet Accuracy and Transferability through AI: A Survey [article]

Chongzhen Zhang, Jianrui Wang, Gary G. Yen, Chaoqiang Zhao, Qiyu Sun, Yang Tang, Feng Qian, Jürgen Kurths
2020 arXiv   pre-print
Firstly, we introduce some basic concepts of transfer learning and then present some preliminaries of adversarial learning, RL and meta-learning.  ...  Finally, we discuss several challenges and future topics for using adversarial learning, RL and meta-learning in autonomous systems.  ...  [236] theoretically analyzed unsupervised domain adaptation re-ID tasks, which bridges the gap between theories of unsupervised domain adaptation and re-ID task.  ... 
arXiv:2003.12948v3 fatcat:qtmjs74p2vh6thdotbhgebdvoi

Domain Adaptation for Reinforcement Learning on the Atari [article]

Thomas Carr, Maria Chli, George Vogiatzis
2018 arXiv   pre-print
We utilise adversarial domain adaptation ideas combined with an adversarial autoencoder architecture.  ...  with the environment to learn a suitable policy.  ...  Adversarial Adaptation methods offer a powerful framework for domain adaptation.  ... 
arXiv:1812.07452v1 fatcat:mdhunianpjdvbn2u6nmqrxn7ta
« Previous Showing results 1 — 15 out of 8,433 results