Filters








24,689 Hits in 5.1 sec

Zero-shot Learning via Simultaneous Generating and Learning [article]

Hyeonwoo Yu, Beomhee Lee
2019 arXiv   pre-print
To overcome the absence of training data for unseen classes, conventional zero-shot learning approaches mainly train their model on seen datapoints and leverage the semantic descriptions for both seen  ...  That is, our network aims to find optimal unseen datapoints and model parameters, by iteratively following the generating and learning strategy.  ...  and in part by the Brain Korea 21 Plus Project.  ... 
arXiv:1910.09446v1 fatcat:ppoylsmjzzgbbbni3doqim2e5e

Model Selection for Generalized Zero-shot Learning [article]

Hongguang Zhang, Piotr Koniusz
2018 arXiv   pre-print
Current approaches combine the auxiliary datapoints and original training data to train the generalized zero-shot learning model and obtain state-of-the-art results.  ...  In the problem of generalized zero-shot learning, the datapoints from unknown classes are not available during training.  ...  An evaluation paper [5] proposes a novel zero-shot learning splits to eliminate the overlap between the classes in zero-shot datasets and ImageNet [5] , and evaluates most popular zero-shot learning  ... 
arXiv:1811.03252v1 fatcat:33ssrcknqvb7nnplfayoaikq5u

Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders [article]

Edgar Schönfeld, Sayna Ebrahimi, Samarth Sinha, Trevor Darrell, Zeynep Akata
2019 arXiv   pre-print
We evaluate our learned latent features on several benchmark datasets, i.e. CUB, SUN, AWA1 and AWA2, and establish a new state of the art on generalized zero-shot as well as on few-shot learning.  ...  Many approaches in generalized zero-shot learning rely on cross-modal mapping between the image feature space and the class embedding space.  ...  Related Work In this section, we present related work on generalized zero-shot learning, few-shot learning and cross-modal reconstruction. Generalized Zero-and Few-Shot Learning.  ... 
arXiv:1812.01784v4 fatcat:px7zvsnsz5a5zfxt7vbvhxnh54

Data-Efficient Pretraining via Contrastive Self-Supervision [article]

Nils Rethmeier, Isabelle Augenstein
2021 arXiv   pre-print
Namely, we analyze: (1) pretraining data (X) efficiency; (2) zero to few-shot label (Y) efficiency; and (3) long-tail generalization, since long-tail preservation has been linked to algorithmic fairness  ...  Transfer learning from high-resource pretraining works well, but research has focused on settings with very large data and compute requirements, while the potential of efficient low-resource learning,  ...  Related Work To simultaneously satisfy the four capability requirements of self-supervised pretraining, long-tail, zero and few-shot learning, CLESS extends on three machine learning concepts (see §3 for  ... 
arXiv:2010.01061v4 fatcat:tml3ujz5ezdxbmn2i2reklyenu

Improving Zero-shot Multilingual Neural Machine Translation for Low-Resource Languages [article]

Chenyang Li, Gongxu Luo
2021 arXiv   pre-print
quality of zero-shot translation, it confronts with two problems: the multilingual NMT model is prone to generate wrong target language when implementing zero-shot translation; the self-learning algorithm  ...  Similarly, it obtains 9.08 and 7.99 BLEU scores on Italian-Romanian zero-shot translation.  ...  It significantly improve the performance of the zero-shot translation and is simultaneously helpful for the low-resource multilingual NMT. • We improve the self-learning method via replacing beam search  ... 
arXiv:2110.00712v1 fatcat:gh3klk27dne7xd5h5qa2xuurxq

A Generative Model for Zero Shot Learning Using Conditional Variational Autoencoders

Ashish Mishra, Shiva Krishna Reddy, Anurag Mittal, Hema A. Murthy
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Zero shot learning in Image Classification refers to the setting where images from some novel classes are absent in the training data but other information such as natural language descriptions or attribute  ...  to an unseen class, we take a different approach and try to generate the samples from the given attributes, using a conditional variational autoencoder, and use the generated samples for classification  ...  Most zero shot learning approaches learn a projection from image space to the class embedding space via a transfer function.  ... 
doi:10.1109/cvprw.2018.00294 dblp:conf/cvpr/MishraRMM18 fatcat:rc2hih2rqjandf3wivveewx4ve

A Generative Model For Zero Shot Learning Using Conditional Variational Autoencoders [article]

Ashish Mishra, M Shiva Krishna Reddy, Anurag Mittal, Hema A Murthy
2018 arXiv   pre-print
Zero shot learning in Image Classification refers to the setting where images from some novel classes are absent in the training data but other information such as natural language descriptions or attribute  ...  to an unseen class, we take a different approach and try to generate the samples from the given attributes, using a conditional variational autoencoder, and use the generated samples for classification  ...  Most zero shot learning approaches learn a projection from image space to the class embedding space via a transfer function.  ... 
arXiv:1709.00663v2 fatcat:6fodcjnxujekra66tf3ckdgqsa

Cross-modal Zero-shot Hashing [article]

Xuanwu Liu, Zhao Li, Jun Wang, Guoxian Yu, Carlotta Domeniconi, Xiangliang Zhang
2019 arXiv   pre-print
Zero-shot hashing (ZSH) aims to learn a hashing model that is trained using only samples from seen categories, but can generalize well to samples of unseen categories.  ...  To address these issues, we propose a general Cross-modal Zero-shot Hashing (CZHash) solution to effectively leverage unlabeled and labeled multi-modality data with different label spaces.  ...  Cross-modal zero-shot hashing can be viewed as a special case of zero-shot learning and hashing learning. Interested readers can refer to [1] , [2] and [25] , [26] .  ... 
arXiv:1908.07388v1 fatcat:uaj7z5dnfrgvvmy7llwjfkuwo4

Attribute-Induced Bias Eliminating for Transductive Zero-Shot Learning [article]

Hantao Yao, Shaobo Min, Yongdong Zhang, Changsheng Xu
2020 arXiv   pre-print
Transductive Zero-shot learning (ZSL) targets to recognize the unseen categories by aligning the visual and semantic information in a joint embedding space.  ...  ., obtaining the 82.8%/75.5%, 97.1%/82.5%, and 73.2%/52.1% for Conventional/Generalized ZSL settings for CUB, AwA2, and SUN datasets, respectively.  ...  Zero-shot learning and Generalized Zero-shot learning, and summarize the results in Table 5 and Table 6 , respectively.  ... 
arXiv:2006.00412v1 fatcat:vpz5hp3otfbdbeyt3aptuj5tla

Zero-Shot Visual Recognition via Bidirectional Latent Embedding [article]

Qian Wang, Ke Chen
2017 arXiv   pre-print
Zero-shot learning for visual recognition, e.g., object and action recognition, has recently attracted a lot of attention.  ...  Unlike most of the existing zero-shot visual recognition methods, we propose a stagewise bidirectional latent embedding framework to two subsequent learning stages for zero-shot visual recognition.  ...  Acknowledgements The authors would like to thank the action editor and all the anonymous reviewers for their invaluable comments that considerably improve the presentation of this manuscript.  ... 
arXiv:1607.02104v4 fatcat:35tawiiuwvgjxhkolsehwstkte

Visual Data Synthesis via GAN for Zero-Shot Video Classification

Chenrui Zhang, Yuxin Peng
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
Most existing methods exploit seento- unseen correlation via learning a projection between visual and semantic spaces.  ...  Zero-Shot Learning (ZSL) in video classification is a promising research direction, which aims to tackle the challenge from explosive growth of video categories.  ...  Acknowledgments This work was supported by National Natural Science Foundation of China under Grant 61771025 and Grant 61532005.  ... 
doi:10.24963/ijcai.2018/157 dblp:conf/ijcai/ZhangP18 fatcat:s46nkoz2dzdfdjqrcm7yi4qytu

Zero-Shot Visual Recognition via Bidirectional Latent Embedding

Qian Wang, Ke Chen
2017 International Journal of Computer Vision  
Zero-shot learning for visual recognition, e.g., object and action recognition, has recently attracted a lot of attention.  ...  Unlike most of the existing zero-shot visual recognition methods, we propose a stagewise bidirectional latent embedding framework to two subsequent learning stages for zero-shot visual recognition.  ...  Acknowledgements The authors would like to thank the action editor and all the anonymous reviewers for their invaluable comments that considerably improve the presentation of this manuscript.  ... 
doi:10.1007/s11263-017-1027-5 fatcat:ofepocl3gvgzbn4y6dwgazmgh4

Zero-Shot Joint Modeling of Multiple Spoken-Text-Style Conversion Tasks using Switching Tokens [article]

Mana Ihori, Naoki Makishima, Tomohiro Tanaka, Akihiko Takashima, Shota Orihashi, Ryo Masumura
2021 arXiv   pre-print
In our proposed zero-shot joint modeling, we switch the individual tasks using multiple switching tokens, enabling us to utilize a zero-shot learning approach to executing simultaneous conversions.  ...  In practice, transcriptions generated by automatic speech recognition systems are not highly readable because they often include many disfluencies and do not include punctuation marks.  ...  The switching tokens enable us to utilize a zero-shot learning approach to executing simultaneous conversions.  ... 
arXiv:2106.12131v1 fatcat:mij3mbqzd5g3dbt36qyr76a4q4

Zero-Shot Kernel Learning

Hongguang Zhang, Piotr Koniusz
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
In this paper, we address an open problem of zero-shot learning.  ...  Despite simplicity of our approach, we obtain state-of-the-art results on several zero-shot learning datasets and benchmarks including a recent AWA2 dataset [45] .  ...  Generalized protocol. Table 3 presents our results on the generalized zero-shot learning protocol.  ... 
doi:10.1109/cvpr.2018.00800 dblp:conf/cvpr/ZhangK18 fatcat:ikyewkgbyregdddazhbjyl65ra

Zero-Shot Kernel Learning [article]

Hongguang Zhang, Piotr Koniusz
2018 arXiv   pre-print
In this paper, we address an open problem of zero-shot learning.  ...  Despite simplicity of our approach, we obtain state-of-the-art results on several zero-shot learning datasets and benchmarks including a recent AWA2 dataset.  ...  Generalized protocol. Table 3 presents our results on the generalized zero-shot learning protocol.  ... 
arXiv:1802.01279v2 fatcat:pzs56n4ihng4bkdl7cnc2i5dri
« Previous Showing results 1 — 15 out of 24,689 results