8,527 Hits in 4.2 sec

Interpreting Stellar Spectra with Unsupervised Domain Adaptation [article]

Teaghan O'Briain, Yuan-Sen Ting, Sébastien Fabbro, Kwang M. Yi, Kim Venn, Spencer Bialek
2020 arXiv   pre-print
We discuss how to achieve mapping from large sets of imperfect simulations and observational data with unsupervised domain adaptation.  ...  Driven by an application to interpret stellar spectroscopic sky surveys, we construct the domain transfer pipeline from two adversarial autoencoders on each domains with a disentangling latent space, and  ...  Thus, in this paper, we propose an unsupervised domain adaptation method, that learns these corrections without human intervention.  ... 
arXiv:2007.03112v1 fatcat:lv3iiv54pbg5hcp43iiu2uqlje

Unsupervised Learned Kalman Filtering [article]

Guy Revach, Nir Shlezinger, Timur Locher, Xiaoyong Ni, Ruud J. G. van Sloun, Yonina C. Eldar
2021 arXiv   pre-print
We also show that we can adapt a pre-trained KalmanNet to changing SS models without providing additional data thanks to the unsupervised capabilities.  ...  With the capability of unsupervised learning, one can use KalmanNet not only to track the hidden state, but also to adapt to variations in the state space (SS) model.  ...  To cope with missing model parameters, data is commonly used for parameter estimation, followed by plugging in the missing parameters into the MB KF and its variants [3, 4] .  ... 
arXiv:2110.09005v1 fatcat:pa63w5m27bhkjcl7jo3lnswf5q

Emerging Trends in Federated Learning: From Model Fusion to Federated X Learning [article]

Shaoxiong Ji and Teemu Saravirta and Shirui Pan and Guodong Long and Anwar Walid
2021 arXiv   pre-print
transfer learning, unsupervised learning, and reinforcement learning.  ...  Federated learning is a new learning paradigm that decouples data collection and model training via multi-party computation and model aggregation.  ...  Fully unsupervised data can be enhanced via domain adaption, where the aim is to transfer knowledge from a labeled domain to an unlabeled one. Taxonomy.  ... 
arXiv:2102.12920v2 fatcat:5fcwfhxibbedbcbuzrfyqdedky

Untranscribed Web Audio for Low Resource Speech Recognition

Andrea Carmantini, Peter Bell, Steve Renals
2019 Interspeech 2019  
For low resource languages, it is difficult to obtain transcribed speech for target domains, while untranscribed data can be collected with minimal effort.  ...  On data from the IARPA MATERIAL programme, our new semi-supervised method outperforms the standard semisupervised method, yielding significant gains when adapting for mismatched bandwidth and domain.  ...  Semi-supervised and lightly supervised adaptation techniques use a base model trained on out-of-domain supervised data to generate targets on in-domain unsupervised data.  ... 
doi:10.21437/interspeech.2019-2623 dblp:conf/interspeech/Carmantini0R19 fatcat:4rqhxle4jrawnotksgjulatyay

Unsupervised Domain Adaptation with a Relaxed Covariate Shift Assumption

Tameem Adel, Han Zhao, Alexander Wong
Domain adaptation addresses learning tasks where training is performed on data from one domain whereas testing is performed on data belonging to a different but related domain.  ...  In addition to the ability to model less restrictive relationships between source and target, modelling can be performed without any target labeled data (unsupervised domain adaptation).  ...  In other frameworks, like Huang et al. (2006) , referred to as unsupervised domain adaptation, the target data is fully unlabeled.  ... 
doi:10.1609/aaai.v31i1.10898 fatcat:wxyxvno7rbaezf3tswoazohgbi

Learning Semantic Representations for Unsupervised Domain Adaptation

Shaoan Xie, Zibin Zheng, Liang Chen, Chuan Chen
2018 International Conference on Machine Learning  
Prior domain adaptation methods address this problem through aligning the global distribution statistics between source domain and target domain, but a drawback of prior methods is that they ignore the  ...  It is important to transfer the knowledge from label-rich source domain to unlabeled target domain due to the expensive cost of manual labeling efforts.  ...  ., 2017) have been explored in domain adaptation. Few-shot domain adaptation considers the task where very few labeled target data are available in training.  ... 
dblp:conf/icml/XieZCC18 fatcat:mizrllhtvzcu3flpvvccnzcype

Unsupervised Domain Adaptation across FMCW Radar Configurations Using Margin Disparity Discrepancy [article]

Rodrigo Hernangomez, Igor Bjelakovic, Lorenzo Servadei, Slawomir Stanczak
2022 arXiv   pre-print
In this work, we consider the problem of unsupervised domain adaptation across radar configurations in the context of deep-learning human activity classification using frequency-modulated continuous-wave  ...  Our experiments extend this technique to radar data, achieving a comparable accuracy to fewshot supervised approaches for the same classification problem.  ...  ACKNOWLEDGMENT We gratefully acknowledge the support of NVIDIA Corporation with the donation of the DGX-1 used for this research.  ... 
arXiv:2203.04588v2 fatcat:575ndjfevrgwnbk5v2hjrftmii

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges

Muhammad Usama, Junaid Qadir, Aunn Raza, Hunain Arif, Kok-lim Alvin Yau, Yehia Elkhatib, Amir Hussain, Ala Al-Fuqaha
2019 IEEE Access  
The focus of this survey paper is to provide an overview of applications of unsupervised learning in the domain of networking.  ...  Recently, there has been a rising trend of employing unsupervised machine learning using unstructured raw network data to improve network performance and provide services, such as traffic engineering,  ...  non-Gaussian features of the data and tries to maximize the fourth moment of linear combination of inputs to extract non-normal source components in the data [99] . 4) NON-NEGATIVE MATRIX FACTORIZATION  ... 
doi:10.1109/access.2019.2916648 fatcat:xutxh3neynh4bgcsmugxsclkna

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges [article]

Muhammad Usama, Junaid Qadir, Aunn Raza, Hunain Arif, Kok-Lim Alvin Yau, Yehia Elkhatib, Amir Hussain, Ala Al-Fuqaha
2017 arXiv   pre-print
The focus of this survey paper is to provide an overview of the applications of unsupervised learning in the domain of networking.  ...  Recently there has been a rising trend of employing unsupervised machine learning using unstructured raw network data to improve network performance and provide services such as traffic engineering, anomaly  ...  correlation with non parametric NN to produce efficient and adaptive results in traffic classification.  ... 
arXiv:1709.06599v1 fatcat:llcg6gxgpjahha6bkhsitglrsm

Unsupervised domain adaptation for the automated segmentation of neuroanatomy in MRI: a deep learning approach [article]

Philip Novosad, Vladimir Fonov, D. Louis Collins
2019 bioRxiv   pre-print
trained using extensive data augmentation with label-preserving transformations which mimic differences between domains.  ...  This work introduces a new method for unsupervised domain adaptation which improves performance in challenging cross-domain applications without requiring any additional annotations on the target domain  ...  of missing labels (rows 1 and 2 of Fig. 6 ).  ... 
doi:10.1101/845537 fatcat:nn4dpm43qrcgjmklarrqfrqpvy

MixStyle Neural Networks for Domain Generalization and Adaptation [article]

Kaiyang Zhou, Yongxin Yang, Yu Qiao, Tao Xiang
2021 arXiv   pre-print
adaptation, with a simple extension to mix feature statistics between labeled and pseudo-labeled instances.  ...  One way to improve domain generalization is to collect diverse source data from multiple relevant domains so that a CNN model is allowed to learn more domain-invariant, and hence generalizable representations  ...  eralization and unsupervised domain adaptation. Mixing vs.  ... 
arXiv:2107.02053v1 fatcat:6hflai4ikbehtldxbq5xxsbvme

Partial Coupling of Optimal Transport for Spoken Language Identification [article]

Xugang Lu, Peng Shen, Yu Tsao, Hisashi Kawai
2022 arXiv   pre-print
In order to reduce domain discrepancy to improve the performance of cross-domain spoken language identification (SLID) system, as an unsupervised domain adaptation (UDA) method, we have proposed a joint  ...  Moreover, since the label of test data is unknown, in the POT, a soft weighting on the coupling based on transport cost is adaptively set during domain alignment.  ...  We first visually check the effect of unsupervised adaptation on language cluster distributions based on the t-Distributed Stochastic Neighbor Embedding (TSNE) [24] .  ... 
arXiv:2203.17036v1 fatcat:djkzm7z76rgzfag5rg4dvhl2yq

Adversarial Feature Augmentation for Unsupervised Domain Adaptation

Riccardo Volpi, Pietro Morerio, Silvio Savarese, Vittorio Murino
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Results show that both enforcing domain-invariance and performing feature augmentation lead to superior or comparable performance to state-of-the-art results in several unsupervised domain adaptation benchmarks  ...  Recent works showed that Generative Adversarial Networks (GANs) can be successfully applied in unsupervised domain adaptation, where, given a labeled source dataset and an unlabeled target dataset, the  ...  Unsupervised domain adaptation.  ... 
doi:10.1109/cvpr.2018.00576 dblp:conf/cvpr/VolpiMSM18 fatcat:o4gf7odzrbb77gwk4mqf7fnccq

Weakly-Supervised Free Space Estimation through Stochastic Co-Teaching

Francois Robinet, Claudia Parera, Christian Hundt, Raphael Frank
2022 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)  
In this work, we propose a novel approach for obtaining free space estimates from images taken with a single roadfacing camera.  ...  In addition, we propose Stochastic Co-Teaching, which is a novel method to select clean samples that leads to enhanced results.  ...  Co-Teaching ensemble) ground truth 0.9360 0.9664 0.9655 Bottom Half no training 0.7550 0.7798 0.9616 Weak Labels [49] no training 0.7900 0.8778 0.8924 Unsupervised Domain Adaptation [20] synthetic data  ... 
doi:10.1109/wacvw54805.2022.00068 fatcat:kcvoxc5d2jdsbfgc3uios5y7m4

An Unsupervised Transfer Learning Framework for Visible-Thermal Pedestrian Detection

Chengjin Lyu, Patrick Heyer, Bart Goossens, Wilfried Philips
2022 Sensors  
To overcome the problem, we propose a novel unsupervised transfer learning framework for multispectral pedestrian detection, which adapts a multispectral pedestrian detector to the target domain based  ...  Dual cameras with visible-thermal multispectral pairs provide both visual and thermal appearance, thereby enabling detecting pedestrians around the clock in various conditions and applications, including  ...  Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s22124416 pmid:35746199 pmcid:PMC9228565 fatcat:qdbvn5bmp5ctfaim2ta43h52fe
« Previous Showing results 1 — 15 out of 8,527 results