Filters








254 Hits in 3.8 sec

Zero-shot multi-label learning via label factorisation

Hang Shao, Yuchen Guo, Guiguang Ding, Jungong Han
2018 IET Computer Vision  
This study considers the zero-shot learning problem under the multi-label setting where each test sample is associated with multiple labels that are unseen in training data.  ...  The authors propose a novel learning framework based on label factorisation for this problem.  ...  Zero-shot multi-label prediction: Solving (3) can provide the model parameters for zero-shot multi-label classification.  ... 
doi:10.1049/iet-cvi.2018.5131 fatcat:5shug5tmyvbqzncizpspcf5toi

Unifying Multi-Domain Multi-Task Learning: Tensor and Neural Network Perspectives [article]

Yongxin Yang, Timothy M. Hospedales
2016 arXiv   pre-print
well as zero-shot domain adaptation (ZSDA), where a model is generated for an unseen domain without any training data.  ...  Moreover, by exploiting the semantic descriptor, it provides neural networks the capability of zero-shot learning (ZSL), where a classifier is generated for an unseen class without any training data; as  ...  X → Z → Y [ or Z + + X / / Y [ Zero-Shot Domain Adaptation Going beyond conventional ZSL, we generalise the notion of zero-shot learning of tasks to zero-shot learning of domains.  ... 
arXiv:1611.09345v1 fatcat:ao45l3bjazcmxmuqyrgeqjw3am

Multi-Task Zero-Shot Action Recognition with Prioritised Data Augmentation [chapter]

Xun Xu, Timothy M. Hospedales, Shaogang Gong
2016 Lecture Notes in Computer Science  
Zero-Shot Learning (ZSL) promises to scale visual recognition by bypassing the conventional model training requirement of annotated examples for every category.  ...  The proposed new model is applied to the challenging zero-shot action recognition problem to demonstrate its advantages over existing ZSL models.  ...  Related Work Zero-Shot Learning Zero-shot Learning (ZSL) [5] aims to generalize existing knowledge to recognize new categories without training examples by re-using a mapping learned from visual features  ... 
doi:10.1007/978-3-319-46475-6_22 fatcat:zmwhktds3vfndj6wh364qpsutu

Zero-Shot Learning and its Applications from Autonomous Vehicles to COVID-19 Diagnosis: A Review

Mahdi Rezaei, Mahsa Shahidi
2020 Intelligence-Based Medicine  
The challenge of learning a new concept, object, or a new medical disease recognition without receiving any examples beforehand is called Zero-Shot Learning (ZSL).  ...  We review over fundamentals and the challenging steps of Zero-Shot Learning, including state-of-the-art categories of solutions, as well as our recommended solution, motivations behind each approach, their  ...  Conclusion In this article, we performed a comprehensive and multi-faceted review on the Zero-Shot/Generalised Zero-shot Learning challenge, its fundamentals, and variants for different scenarios and applications  ... 
doi:10.1016/j.ibmed.2020.100005 pmid:33043311 pmcid:PMC7531283 fatcat:qzyaf7gpufhermyg5gvank5cja

Zero-Shot Learning and its Applications from Autonomous Vehicles to COVID-19 Diagnosis: A Review

Mahdi Rezaei, Mahsa Shahidi
2020 Social Science Research Network  
A B S T R A C T The challenge of learning a new concept, object, or a new medical disease recognition without receiving any examples beforehand is called Zero-Shot Learning (ZSL).  ...  We review over fundamentals and the challenging steps of Zero-Shot Learning, including state-of-the-art categories of solutions, as well as our recommended solution, motivations behind each approach, their  ...  Conclusion In this article, we performed a comprehensive and multi-faceted review on the Zero-Shot/Generalised Zero-shot Learning challenge, its fundamentals, and variants for different scenarios and applications  ... 
doi:10.2139/ssrn.3624379 fatcat:yifnxv46rjf6pgndowkxzmo5o4

Zero-Shot Learning and its Applications from Autonomous Vehicles to COVID-19 Diagnosis: A Review [article]

Mahdi Rezaei, Mahsa Shahidi
2020 arXiv   pre-print
The challenge of learning a new concept, object, or a new medical disease recognition without receiving any examples beforehand is called Zero-Shot Learning (ZSL).  ...  In this review paper, we present the definition of the problem, we review over fundamentals, and the challenging steps of Zero-Shot Learning, including state-of-the-art categories of solutions as well  ...  One-Shot learning (OSL) and Few-shot learning (FSL) are two solutions that are able to learn new categories via one or a few images, respectively [1] , [2] , [3] .  ... 
arXiv:2004.14143v2 fatcat:erh6xyog7bb5vofcebkk2zxumm

Deep Multi-task Representation Learning: A Tensor Factorisation Approach [article]

Yongxin Yang, Timothy Hospedales
2017 arXiv   pre-print
Our approach is based on generalising the matrix factorisation techniques explicitly or implicitly used by many conventional MTL algorithms to tensor factorisation, to realise automatic learning of end-to-end  ...  Most contemporary multi-task learning methods assume linear models. This setting is considered shallow in the era of deep learning.  ...  METHODOLOGY PRELIMINARIES We first recap some tensor factorisation basics before explaining how to factorise DNN weight tensors for multi-task representation learning.  ... 
arXiv:1605.06391v2 fatcat:n5wqvr347nenzdr2sopcavoguu

Towards Universal Representation for Unseen Action Recognition [article]

Yi Zhu, Yang Long, Yu Guan, Shawn Newsam, Ling Shao
2018 arXiv   pre-print
We first address UAR as a Generalised Multiple-Instance Learning (GMIL) problem and discover 'building-blocks' from the large-scale ActivityNet dataset using distribution kernels.  ...  Hence, word vectors have been preferred for zero-shot action recognition, since only category names are required for constructing the label embeddings.  ...  Zero-shot action recognition has recently drawn considerable attention because of its ability to recognize unseen action categories without any labelled examples.  ... 
arXiv:1803.08460v1 fatcat:becturjwxfhtxkna4hdrw5436i

Towards Universal Representation for Unseen Action Recognition

Yi Zhu, Yang Long, Yu Guan, Shawn Newsam, Ling Shao
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We first address UAR as a Generalised Multiple-Instance Learning (GMIL) problem and discover 'building-blocks' from the large-scale ActivityNet dataset using distribution kernels.  ...  Hence, word vectors have been preferred for zero-shot action recognition, since only category names are required for constructing the label embeddings.  ...  Zero-shot action recognition has recently drawn considerable attention because of its ability to recognize unseen action categories without any labelled examples.  ... 
doi:10.1109/cvpr.2018.00983 dblp:conf/cvpr/ZhuLGN018 fatcat:sfpjwv3l2bbarj3xjatb4swfri

Multi-Facet Clustering Variational Autoencoders [article]

Fabian Falck, Haoting Zhang, Matthew Willetts, George Nicholson, Christopher Yau, Chris Holmes
2021 arXiv   pre-print
, that learns multiple clusterings simultaneously, and is trained fully unsupervised and end-to-end.  ...  In this paper, we introduce Multi-Facet Clustering Variational Autoencoders (MFCVAE), a novel class of variational autoencoders with a hierarchy of latent variables, each with a Mixture-of-Gaussians prior  ...  B.2 Multi-Facet VaDE Trick (factorised distribution) Proof.  ... 
arXiv:2106.05241v2 fatcat:pknxdvuhc5dztp3cb5zc6stgve

Robust Task Clustering for Deep Many-Task Learning [article]

Mo Yu, Xiaoxiao Guo, Jinfeng Yi, Shiyu Chang, Saloni Potdar, Gerald Tesauro, Haoyu Wang, Bowen Zhou
2018 arXiv   pre-print
We investigate task clustering for deep-learning based multi-task and few-shot learning in a many-task setting.  ...  Our task clustering approach also extends metric-based few-shot learning methods to adapt multiple metrics, which demonstrates empirical advantages when the tasks are diverse.  ...  We propose a task-clustering framework for both multi-task learning (MTL) and few-shot learning (FSL) settings.  ... 
arXiv:1708.07918v2 fatcat:iladbqiqnnbvxby4k2fbk7d73e

Snowflake: Scaling GNNs to High-Dimensional Continuous Control via Parameter Freezing [article]

Charlie Blake, Vitaly Kurin, Maximilian Igl, Shimon Whiteson
2022 arXiv   pre-print
Recent research has shown that graph neural networks (GNNs) can learn policies for locomotion control that are as effective as a typical multi-layer perceptron (MLP), with superior transfer and multi-task  ...  A key motivation for the use of GNNs in the supervised learning setting is their applicability to large graphs, but this benefit has not yet been realised for locomotion control.  ...  NERVENET achieves multi-task and transfer learning across morphologies, even in the zero-shot setting (i.e., without further training), which standard MLP-based policies fail to achieve.  ... 
arXiv:2103.01009v3 fatcat:6rz5r2m3eneornrtdihtgz6vei

AttKGCN: Attribute Knowledge Graph Convolutional Network for Person Re-identification [article]

Bo Jiang, Xixi Wang, Jin Tang
2019 arXiv   pre-print
In this paper, we propose to model these attribute dependencies via a novel attribute knowledge graph (AttKG), and propose a novel Attribute Knowledge Graph Convolutional Network (AttKGCN) to solve Re-ID  ...  Recently, attributes have been demonstrated beneficially in guiding for learning more discriminative feature representations for Re-ID.  ...  Recently, knowledge graph convolutional networks have been developed for zero-shot learning and multi-label recognition. Wang et al.  ... 
arXiv:1911.10544v1 fatcat:nrhackccljcvbiriaje4bajxvm

Massively Multilingual Transfer for NER [article]

Afshin Rahimi, Yuan Li, Trevor Cohn
2019 arXiv   pre-print
We propose two techniques for modulating the transfer, suitable for zero-shot or few-shot learning, respectively.  ...  Low-resource named entity recognition via multi-source projection: Not quite there yet?  ...  H Zero-Shot Transfer as Truth Inference One way to improve the performance of the ensemble system is to select a subset of component models carefully, or more generally, learn a non-uniform weighting  ... 
arXiv:1902.00193v4 fatcat:y2rjryqrjnbs3asksdi3t2ardy

Multimodal One-shot Learning of Speech and Images

Ryan Eloff, Herman A. Engelbrecht, Herman Kamper
2019 ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
Meanwhile, state-of-the-art machine learning models-which aim to challenge these human learning abilities-require large amounts of labelled training data to enable successful generalisation.  ...  Specifically, we consider spoken word learning with co-occurring visual context in a one-shot setting, where an agent must learn novel concepts (words and object categories from a single joint audio-visual  ...  Part II • In Chapter 4, we introduce a framework for multimodal one-shot learning relying on indirect matching via unimodal comparisons.  ... 
doi:10.1109/icassp.2019.8683587 dblp:conf/icassp/EloffEK19 fatcat:47yfbmhsg5bbbdeiivcglj3vtu
« Previous Showing results 1 — 15 out of 254 results