150,587 Hits in 4.1 sec

On the Effectiveness of Discretizing Quantitative Attributes in Linear Classifiers

Nayyar A. Zaidi, Yang Du, Geoffrey I. Webb
2020 IEEE Access  
LR d means a logistic regression classifier trained on discretized data (i.e., quantitative attributes are converted into qualitative ones).  ...  In the following, LR means a logistic regression classifier trained with original data (i.e., with both qualitative and quantitative attributes).  ... 
doi:10.1109/access.2020.3034955 fatcat:h5xccmiiy5gc5hxgd3sivcuhpm

On the Effectiveness of Discretizing Quantitative Attributes in Linear Classifiers [article]

Nayyar A. Zaidi, Yang Du, Geoffrey I. Webb
2017 arXiv   pre-print
It is often motivated by the limitation of some learners to qualitative data.  ...  This motivates a systematic study of the effectiveness of discretizing quantitative attributes for other linear classifiers.  ...  We denote linear classifier optimizing the conditional log-likelihood as LR, a linear classifier optimizing the Hinge loss as SVC (support vector classifier) and a linear classifier optimizing the mean-square-error  ... 
arXiv:1701.07114v1 fatcat:jqh6maonmbh7vkoxdzaqk5q2ei

Assessing glaucoma in retinal fundus photographs using Deep Feature Consistent Variational Autoencoders [article]

Sayan Mandal, Alessandro A. Jammal, Felipe A. Medeiros
2021 arXiv   pre-print
Latent representations of size as low as 128 from our model got a 0.885 area under the receiver operating characteristic curve when trained with Support Vector Classifier.  ...  Deep learning methods have partially solved this dilemma by bypassing the marker identification stage and analyzing high-level information directly to classify the data.  ...  Figure 9 . 9 5-Fold Cross Validation Performance Trends for optimal SVC classifier model trained with DFC-VAE latent representations obtained at different latent space sizes.  ... 
arXiv:2110.01534v1 fatcat:pc6vt4giljb4tagjq7gozktn2u

Classification Representations Can be Reused for Downstream Generations [article]

Saisubramaniam Gopalakrishnan, Pranshu Ranjan Singh, Yasin Yazici, Chuan-Sheng Foo, Vijay Chandrasekhar, ArulMurugan Ambikapathi
2020 arXiv   pre-print
classifier for the downstream task of sample generation.  ...  Interestingly, such representations allow for controlled sample generations for any given class from existing samples and do not require enforcing prior distribution.  ...  For sake of clarity, let f cls (x; θ, W, b) be the classifier that maps the data sample x to class label y.  ... 
arXiv:2004.07543v1 fatcat:jheawxxqyrbn3jwv4imdo6c2wq

Activation Maximization with a Prior in Speech Data

Sho Inoue, Tad Gonsalves
2021 American Journal of Computer Science and Technology  
In AM, the input data is optimized to find the data that activates the selected neuron.  ...  the most suitable condition for AM in audio domain data.  ...  The classifier, which is essential for AM, varied in the structure and representation of data.  ... 
doi:10.11648/j.ajcst.20210403.13 fatcat:lvf4k2yvr5azdm46d4k4j56ac4

Page 584 of Journal of Cognitive Neuroscience Vol. 17, Issue 4 [page]

2005 Journal of Cognitive Neuroscience  
Optimal classifiers were created for all possible pairs of object categories. These classifiers operated on the PC coordinates derived from the stimuli presented to subjects in the fMRI experi- ment.  ...  These findings are also consistent with the previous data showing that the preferred regions for faces and houses may be impaired at classifying nonpreferred objects (Spiridon & Kanwisher, 2002).  ... 

Learning the Matching Function [article]

Ľubor Ladický, Christian Häne, Marc Pollefeys
2015 arXiv   pre-print
Proposed classifier gives reliable estimations of pixel disparities already without any form of regularization.  ...  The matching function for the problem of stereo reconstruction or optical flow has been traditionally designed as a function of the distance between the features describing matched pixels.  ...  methods for the corresponding optimization problem.  ... 
arXiv:1502.00652v1 fatcat:s5umg47ph5ckdm2vwriryyo4vm

Learning 3D joint constraints from vision-based motion capture datasets

Pramod Murthy, Hammad T. Butt, Sandesh Hiremath, Alireza Khoshhal, Didier Stricker
2019 IPSJ Transactions on Computer Vision and Applications  
We validate precision-accuracy trade-off of discriminators and qualitatively evaluate classified poses with an interactive tool.  ...  A data-driven approach is used to learn human joint limits from 3D motion capture datasets.  ...  would like to acknowledge the support provided by Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI), Kaiserslautern, Germany and National University of Sciences of Technology (NUST), Pakistan for  ... 
doi:10.1186/s41074-019-0057-z fatcat:gp3jlddc4vcjnfz3siqx272oom

Universal Spectral Adversarial Attacks for Deformable Shapes [article]

Arianna Rampini, Franco Pestarini, Luca Cosmo, Simone Melzi, Emanuele Rodolà
2021 arXiv   pre-print
Our attacks are universal, in that they transfer across different shapes, different representations (meshes and point clouds), and generalize to previously unseen data.  ...  ., unique perturbations that transfer across different data points) has only been demonstrated for images to date.  ...  Acknowledgments We gratefully acknowledge Luca Moschella for the technical help.  ... 
arXiv:2104.03356v1 fatcat:wrdemn4qw5gcxfyx4rysznl5eq

Discriminatively Trained Dense Surface Normal Estimation [chapter]

L'ubor Ladický, Bernhard Zeisl, Marc Pollefeys
2014 Lecture Notes in Computer Science  
We apply our method to two challenging data sets containing images of man-made environments, the indoor NYU2 data set and the outdoor KITTI data set.  ...  In this work we propose the method for a rather unexplored problem of computer vision -discriminatively trained dense surface normal estimation from a single image.  ...  Qualitative results of our method on the KITTI data set. The ground truth colours are the same as for the NYU2 data set.  ... 
doi:10.1007/978-3-319-10602-1_31 fatcat:4rxoi7v33zc3leqz6z4x2vw2yq

PSO with mutation for fuzzy classifier design

C. Rania, S.N. Deepa
2010 Procedia Computer Science  
This paper presents a hybrid Particle Swarm Optimization based approach for fuzzy classifier design which incorporates the concept of mutation from evolutionary computations.  ...  The performance of the proposed MutPSO approach is demonstrated through development of fuzzy classifiers for iris data available in UCI machine learning repository.  ...  the performance of the classifier for the Iris data are explained here.  ... 
doi:10.1016/j.procs.2010.11.040 fatcat:4wdsc4nme5fizaqejgf7mbv22a

Representation Learning via Adversarially-Contrastive Optimal Transport [article]

Anoop Cherian, Shuchin Aeron
2020 arXiv   pre-print
In this paper, we study the problem of learning compact (low-dimensional) representations for sequential data that captures its implicit spatio-temporal cues.  ...  To generate the adversarial distribution, we propose a novel framework connecting Wasserstein GANs with a classifier, allowing a principled mechanism for producing good negative distributions for contrastive  ...  Specifically, our classifier consists of a single FCN(d, c), where c denotes the number of data classes.  ... 
arXiv:2007.05840v1 fatcat:5xb5skt43bbjnezhopgshflzwe

Survey on Tea Discriminator

Princee Gupta, Rajesh K.Shukla
2015 International Journal of Computer Applications  
the qualitative and quantitative composition of sample and complex solutions.  ...  The optimization of sample preparation, signal processing, feature extraction, classifier are as important as choice of sensors within the array in enhancing the performance of the organoleptic system  ...  Impedance tongue are sensor array for qualitative and quantitative analysis and it is used to differentiate basic standard taste.  ... 
doi:10.5120/21760-4996 fatcat:d6q3hmd24rcwfkyi7mkldllo6u

Contrast Phase Classification with a Generative Adversarial Network [article]

Yucheng Tang, Ho Hin Lee, Yuchen Xu, Olivia Tang, Yunqiang Chen, Dashan Gao, Shizhong Han, Riqiang Gao, Camilo Bermudez, Michael R. Savona, Richard G. Abramson, Yuankai Huo, Bennett A. Landman
2019 arXiv   pre-print
Previous studies with deep learning frameworks have been proposed for classifying contrast enhancement with networks inspired by computer vision.  ...  To capture and compensate for the complex contrast changes, we propose a novel discriminator in the form of a multi-domain disentangled representation learning network.  ...  We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X Pascal GPU used for this research.  ... 
arXiv:1911.06395v1 fatcat:nulidrv3jvgajjprpibcl6ttta

Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data

O.V. Novytskyi, Institute of Software Systems NAS of Ukraine
Ontology-based knowledge representation models in the context of big data are one way to reduce complexity for data processing across methods of semantic description.  ...  The proper management of ontology-based knowledge representation models through offered methods and techniques brings improved data integration, big data quality, and business process integration.  ...  The reasoner makes it possible to classify even big data knowledgebase and complex information for minutes.  ... 
doi:10.15407/pp2021.04.019 fatcat:qmn4howe2bcjri3fxjpmuhea34
« Previous Showing results 1 — 15 out of 150,587 results