Filters








42,054 Hits in 4.3 sec

Partial Least Squares Regression Performs Well in MRI-Based Individualized Estimations

Chen Chen, Xuyu Cao, Lixia Tian
2019 Frontiers in Neuroscience  
(multi-label learning).  ...  importance for these estimations.  ...  on multi-label learning (R = 0.536, compared to R = 0.525 for single-label learning).  ... 
doi:10.3389/fnins.2019.01282 pmid:31827420 pmcid:PMC6890557 fatcat:v42hqfd5mrffheumamtlo3dcbi

Multi-task Learning Using Multi-modal Encoder-Decoder Networks with Shared Skip Connections

Ryohei Kuga, Asako Kanezaki, Masaki Samejima, Yusuke Sugano, Yasuyuki Matsushita
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
Multi-task learning is a promising approach for efficiently and effectively addressing multiple mutually related recognition tasks.  ...  learning.  ...  In contrast to such multi-modal singletask learning methods, relatively few studies have been made on multi-modal multi-task learning. Ehrlich et al.  ... 
doi:10.1109/iccvw.2017.54 dblp:conf/iccvw/KugaKSSM17 fatcat:b6er3vhppbfi7fxorcy5xawnhi

Multi-Instance Dynamic Ordinal Random Fields for Weakly Supervised Facial Behavior Analysis

Adria Ruiz, Ognjen Rudovic, Xavier Binefa, Maja Pantic
2018 IEEE Transactions on Image Processing  
We propose a Multi-Instance-Learning (MIL) approach for weakly-supervised learning problems, where a training set is formed by bags (sets of feature vectors or instances) and only labels at bag-level are  ...  To this end, we propose Multi-Instance Dynamic Ordinal Random Fields (MI-DORF). In this framework, we treat instance-labels as temporally-dependent latent variables in an Undirected Graphical Model.  ...  In order to learn the model F from T , it is necessary to incorporate prior knowledge defining the Multi-Instance relation between labels y and latent ordinal states h.  ... 
doi:10.1109/tip.2018.2830189 pmid:29993690 fatcat:bryu22dktrgxzcbswy6eesqc7y

Overcoming data scarcity with transfer learning [article]

Maxwell L. Hutchinson, Erin Antono, Brenna M. Gibbons, Sean Paradiso, Julia Ling, Bryce Meredig
2017 arXiv   pre-print
Here, we describe and compare three techniques for transfer learning: multi-task, difference, and explicit latent variable architectures.  ...  For activation energies of steps in NO reduction, the explicit latent variable method is not only the most accurate, but also enjoys cancellation of errors in functions that depend on multiple tasks.  ...  Unlike multi-task learning and explicit latent variables, difference learning cannot be used directly for multi-class classification.  ... 
arXiv:1711.05099v1 fatcat:nbk535l4cfbtjgwn5xyfckmxse

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.  ...  This is achieved by establishing a mapping connecting low-level features and a semantic description of the label space, referred as visual-semantic mapping, on auxiliary data.  ...  Kullback-Leibler Importance Estimation Procedure (KLIEP) We first introduce the way to estimate a per-instance auxiliary-data weight given the distribution of target data X te .  ... 
doi:10.1007/978-3-319-46475-6_22 fatcat:zmwhktds3vfndj6wh364qpsutu

Knowledge Transfer for Multi-labeler Active Learning [chapter]

Meng Fang, Jie Yin, Xingquan Zhu
2013 Lecture Notes in Computer Science  
In this paper, we address multi-labeler active learning, where data labels can be acquired from multiple labelers with various levels of expertise.  ...  To solve this problem, we propose a new probabilistic model that transfers knowledge from a rich set of labeled instances in some auxiliary domains to help model labelers' expertise for active learning  ...  To the best of our knowledge, our work is the first to leverage transfer learning to help model labelers' expertise for multi-labeler active learning problem.  ... 
doi:10.1007/978-3-642-40988-2_18 fatcat:dfgsjkoisjb5zmzsm62o3z5kh4

Cross-Domain Multitask Learning with Latent Probit Models [article]

Shaobo Han, Lawrence Carin
2012 arXiv   pre-print
We derive theoretical bounds for the estimation error of the classifier in terms of the sparsity of domain transforms. An expectation-maximization algorithm is derived for learning the LPM.  ...  We assume the data in multiple tasks are generated from a latent common domain via sparse domain transforms and propose a latent probit model (LPM) to jointly learn the domain transforms, and the shared  ...  Introduction There are two basic approaches for analysis of data from two or more tasks, single-task learning (STL) and multi-task learning (MTL).  ... 
arXiv:1206.6419v1 fatcat:4ggiwwge2fd3hhggj4zylzumtq

To Avoid the Pitfall of Missing Labels in Feature Selection: A Generative Model Gives the Answer

Yuanyuan Xu, Jun Wang, Jinmao Wei
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In multi-label learning, instances have a large number of noisy and irrelevant features, and each instance is associated with a set of class labels wherein label information is generally incomplete.  ...  These missing labels possess two sides like a coin; people cannot predict whether their provided information for feature selection is favorable (relevant) or not (irrelevant) during tossing.  ...  ., N k (i)) plays an important role in estimating the label observability.  ... 
doi:10.1609/aaai.v34i04.6127 fatcat:c7mwfflsozckdl7wtwpm3pqmha

User Satisfaction Estimation with Sequential Dialogue Act Modeling in Goal-oriented Conversational Systems [article]

Yang Deng, Wenxuan Zhang, Wai Lam, Hong Cheng, Helen Meng
2022 arXiv   pre-print
In this paper, we propose a novel framework, namely USDA, to incorporate the sequential dynamics of dialogue acts for predicting user satisfaction, by jointly learning User Satisfaction Estimation and  ...  User Satisfaction Estimation (USE) is an important yet challenging task in goal-oriented conversational systems.  ...  ACKNOWLEDGMENTS This research/paper was supported by the Center for Perceptual and Interactive Intelligence (CPII) Ltd under the Innovation and Technology Commission's InnoHK scheme.  ... 
arXiv:2202.02912v1 fatcat:a7i2amxrjfh4rezbhinpjwd47y

Combining Generative/Discriminative Learning for Automatic Image Annotation and Retrieval

Zhixin Li, Zhenjun Tang, Weizhong Zhao, Zhiqing Li
2012 International Journal of Intelligence Science  
Furthermore, we propose a hybrid framework which employs continuous PLSA to model visual features of images in generative learning stage and uses ensembles of classifier chains to classify the multi-label  ...  Since the framework combines the advantages of generative and discriminative learning, it can predict semantic annotation precisely for unseen images.  ...  Parameters Setting An important parameter of the experiment is the number of latent aspects for the PLSA-based models.  ... 
doi:10.4236/ijis.2012.23008 fatcat:y7mhhjarejhlpleycxh6fsdwuu

Multi-space Variational Encoder-Decoders for Semi-supervised Labeled Sequence Transduction [article]

Chunting Zhou, Graham Neubig
2017 arXiv   pre-print
In this paper we propose multi-space variational encoder-decoders, a new model for labeled sequence transduction with semi-supervised learning.  ...  The generative model can use neural networks to handle both discrete and continuous latent variables to exploit various features of data.  ...  Acknowledgments The authors thank Jiatao Gu, Xuezhe Ma, Zihang Dai and Pengcheng Yin for their helpful discussions. This work has been supported in part by an Amazon Academic Research Award.  ... 
arXiv:1704.01691v2 fatcat:5w6gxzdg45b4piiogt3xv2gmla

SceneCode: Monocular Dense Semantic Reconstruction using Learned Encoded Scene Representations [article]

Shuaifeng Zhi, Michael Bloesch, Stefan Leutenegger, Andrew J. Davison
2019 arXiv   pre-print
label estimates for each surface element (depth pixels, surfels, or voxels).  ...  Using this learned latent space, we can tackle semantic label fusion by jointly optimising the low-dimenional codes associated with each of a set of overlapping images, producing consistent fused label  ...  we can use the learned latent space to integrate multi-view While [3] encoded only geometry, here we show that we semantic labels, and build a monocular dense SLAM sys- can extend the same conditional  ... 
arXiv:1903.06482v2 fatcat:3sjl5x3ovzhadgodxaul4fv47a

Multi-space Variational Encoder-Decoders for Semi-supervised Labeled Sequence Transduction

Chunting Zhou, Graham Neubig
2017 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
In this paper we propose multi-space variational encoderdecoders, a new model for labeled sequence transduction with semi-supervised learning.  ...  The generative model can use neural networks to handle both discrete and continuous latent variables to exploit various features of data.  ...  Acknowledgments The authors thank Jiatao Gu, Xuezhe Ma, Zihang Dai and Pengcheng Yin for their helpful discussions. This work has been supported in part by an Amazon Academic Research Award.  ... 
doi:10.18653/v1/p17-1029 dblp:conf/acl/ZhouN17 fatcat:6u4b6fex5fflzo4mhlu6j44chq

Large-Scale Bayesian Multi-Label Learning via Topic-Based Label Embeddings

Piyush Rai, Changwei Hu, Ricardo Henao, Lawrence Carin
2015 Neural Information Processing Systems  
We present a scalable Bayesian multi-label learning model based on learning lowdimensional label embeddings.  ...  This makes the model particularly appealing for real-world multi-label learning problems where the label matrix is usually very massive but highly sparse.  ...  Finally, although not a focus of this paper, some other important aspects of the multi-label learning problem have also been looked at in recent work.  ... 
dblp:conf/nips/RaiHHC15 fatcat:4esjlcdeh5hrddm7v27fuoq72e

Learning Disentangled Representations with Semi-Supervised Deep Generative Models [article]

N. Siddharth, Brooks Paige, Jan-Willem van de Meent, Alban Desmaison, Noah D. Goodman, Pushmeet Kohli, Frank Wood, Philip H.S. Torr
2017 arXiv   pre-print
We further define a general objective for semi-supervised learning in this model class, which can be approximated using an importance sampling procedure.  ...  for the remaining variables.  ...  This parameter controls the relative weight of the labelled examples relative to the unlabelled examples in the data.  ... 
arXiv:1706.00400v2 fatcat:havnuwvn65gx7k6t5jv2xaivaa
« Previous Showing results 1 — 15 out of 42,054 results